I mean that, for example, sometimes I don't think "what are my opponent's threats" even I know that I have to do, instead I start to think to my plan
How to correct your tought process
2 important parts of a good thought process are knowledge and calculation, so my advice would be to start with training your calculation (excersises) and augmenting your knowledge (no idea how to that gradually from your level).

Pdve,
I am not being disagreeable for the sake of it, but when you say
"everyone has a unique thought process",
I do not think this is necessarily true for much of Chess. In any case, it does not help people. Most of us had the benefit of a good education, where we were first taught the basics in each subject (a deficiency in "modern education" theory, which over-emphasizes the self-esteem of the child to the detriment of their actual learning), so it seems only consistent that we try to help other with sound advice, especially when they ask for such advice.
For example, there are guiding principles for the Opening phase of a Chess game - concerning piece development, King safety and connecting Rooks - which everyone should heed, whatever their first few opening moves.
Likewise, there are guidelines for the Middle game, for example, as to whether to focus on a King-side, central or Queen-side attack, based on such considerations as your piece development and pawn structure, and what your opponent seems to be doing. The idea of "busting through" with a pawn move, where you have prepared more attackers than your opponent has defenders - these are also important ideas. Likewise, the principles of "when" and "when not" it is better to trade Queens or other pieces - "who benefits from a Queen trade?", " do I want a more open or closed game?", "can I swap my bad bishop for his good bishop".
And finally, there are guidelines for the Endgame, such as Rooks belong behind passed pawns (both yours if trying to promote, and your opponent's if you are tring to prevent promotion), the King should be placed in front of a pawn to shoulder off the opponent King in K+p v K endgames, etc. etc.
As you, yourself have gone through Chess Mentor, you may (like me) have been in awe with players like Tal, with their spectacular multiple sacrifices (including of the Queen) to strip the opponent's King of all defence, and then move in for the kill. But did you see them violate Chess principles? I think not.

ProfBlunderer,
You say:
I mean that, for example, sometimes I don't think "what are my opponent's threats" even I know that I have to do, instead I start to think to my plan
Here are my notes from studying Josh Waitzkin’s Academy in Chessmaster (still useful software to have, although no longer supported by Ubisoft). I might update some notes, but they are still sound.
Check list before you move or commense a series of moves:
1. 1. Before you move, evaluate material equality and inequality - both quantity (pawn equivalents) and quality (a locked in, bad bishop is worth zero, until it is activated)
2. Look at the whole board – what are potential piece moves, along files, ranks and diagonals? Evaluate their tactic or strategic potential
3. Evaluate tactical or strategic “pawn value” of hanging or unprotected pieces - own & opponents
4. Check for pieces that can be captured now – own & opponents (but do not become overly materialistic)
5. Check for pieces actually or potentially trapped – own & opponents
6. Evaluate moves that put the king in check – own & opponent’s, traps, escape routes
7. Look for “zug wang” or self-destruction potential, where any move will hurt – own & opponents
8. What you would like to do (or where to go) to attack or defend. Ditto opponent’s escape routes or threats
Your Arsenal:
9. Remove the defender of a key attacking square via sacrifice, threatened capture
10. Forks – actual or potential (achievable in 1+ moves); e.g. Knight forks, pawn advances forking 2 pieces, Queen settles next to contiguous Knight and Bishop, etc.
Beware “illusory forks” where forked pieces can support each other, or escape with a King check
11. Skewers – R>>K>>Q, where Q is more valuable
12. Pins – p>>B>>Q. Beware “illusory pins”, where pinned piece can attack and defend simultaneously by deft repositioning
13. Discovered attacks- B+p^>>K>>Q (p^ = pawn moves out of way)
14. Double attacks and king checks- B>>K<<R, Rw>>Kb<<Kw
15. Evaluate potential and value of king checks
16. Look for FORCE and CONTINUATIONS; e.g. king checks, arsenal apps. and captures or mate
17. Look for opportunity to immediately checkmate, rather than capture a piece (the Queen); e.g. attacker can check the King and the escape route is blocked or can be attacked next
18. AGAIN, evaluate what is the best move or tactic. There may be a better move than a simple capture
Rooks
19. 19. Rooks prefer open files or ranks, unobstructed by own pawns
20. If 2 open files, place Rook on open file away from King, and then bring King to defend near open file
21. Rooks (almost always) belong behind passed pawns, whether yours or an opponent’s
22. Note on back rank, queens and knights can intervene between defender rook and queening square
Bishops
23. 23. Bishops are long-range attackers, with open diagonals, and usually prefer OPEN games where the central pawn structure is not blocked or grid-locked
24. “Good” bishops have their own pawns on opposite coloured squares (their movement is not blocked) and their opponent’s pawns and pieces are on same coloured squares (many potential targets)
25. “Bad” bishops “ have their own pawns on same coloured squares (their movement is impeded) and their opponent’s pawns and pieces are on opposite coloured squares (few potential targets)
26. If you have a “bad” bishop (few targets), consider a trade with an opponent’s “good” bishop to improve your overall position. Or maybe a pawn trade will allow your “bad” bishop to become unblocked
27. 2 bishops in unison are usually better than 1 bishop and 1 knight
28. Bishops will usually increase in value as game opens up, and especially in end-game (particularly if bishop attacks a queening square for a pawn seeking promotion)
Knights
29. 29. Knights are usually better in CLOSED games because they can leap over other pieces
30. Knights should occupy central positions, where they overlook 8 squares, and are close to or in short range of the action, and can readily attack or fork enemy pieces, or participate in a smothered mate
31. We want to find “outposts” for a knight, a central square in locked or stable pawn formations, deep in enemy territory, protected by our pawns and where it cannot be easily attacked by an enemy’s pawns
32. Central knights hit 8 squares, whereas rim knights hit 4 and corner nights hit 2
33. Rim and corner knights are susceptible to being trapped by a bishop of the opposite colour
34. Knight alone can prevent King and pawn from queening safely, unlike King and bishop against the solitary bishop
Queen
35. 35. Queen is so powerful, it does not need a particular pawn structure to find her course of action
36. Queen is a powerful long-range attacker and usually prefers open games; but can also operate in closed games
37. Queen combines powers of Rook and Bishop
38. Can create powerful forks, simultaneously attacking the rectangular position of an enemy Bishop and diagonal position of a Rook or King
39. Can participate in Q-R, R-Q, R-Q-R, B-Q and Q-B batteries
40. Usually a waste to use the Queen to blockade a passed pawn. Devise an alternative.
Although that list may seem like a lot to remember, a lot is commonsense or plain logical, and the rest you can assimilate naturally.
How to correct your tought process?
I noticed you misspelled the word "thought" in the title of this thread. There might be a clue there. Just a tought.
It might help to take time to review your process to determine if it conforms to the standard you have in mind before you press send or move a piece.
Killing! :)