P.S. The selection of candidate moves is where you whittle down the move choices to a manageable load, since you can't analyze every possible move and probably don't want to.
The selection of candidate moves is an art that will develop as your tactical vision, strategic vision, and board vision (what pieces really do, how they can really move, and who really has the advantage, in a given position) develop with experience.
Note that keeping a journal and using a thinking process concentrates and accelerates that experience, since on each move you are in essence solving a chess problem, and are often working out, not just one, but multiple lines of tactics. You're also evaluating positions more deeply than the typical wood-pusher.
That said, candidate moves should focus on three basic types of moves:
(1) Forcing moves, which require some particular reply from your opponent (or one of a small set of replies). This makes your opponent more predictable by restricting his options, and tends to give you the initiative by allowing you to move the game in a direction of your choosing, not his. It also makes analysis of candidate moves more fruitful, since non-forcing moves allow so many possible replies that in just a few ply the probability of any line you examine actually occurring is close to zero; and what is the point of spending a lot of time analyzing lines that will probably never occur?
(2) Moves that restrict your opponent's degrees of freedom (i.e., which restrict the mobility of his pieces). Sometimes this is a type of forcing move also, if it sets him up for a future forcing move or channels his reply into lines more to your liking.
(3) Moves that develop/activate your own pieces and/or insure king safety. Combinations involve multiple pieces, so development is important though not the be-all and end-all.
How you should view your journal:
Your journal should be a tool subservient to the development of an organized thinking process. Weak players tend to make one or more of the following errors:
(1) Moving without considering your opponent's pre-existing checks, captures and threats. You shouldn't react purely defensively if you can respond to a threat with a better threat, but you do have to consider what your opponent can do after his move, and not just be wrapped up in your own plans, oblivious. Chess coach Dan Heisman calls this kind of blind playing "coin-flip chess".
(2) Moving without considering how your opponent can respond to your own move (i.e., his checks, captures and threats). This sounds like the first mistake, but that involved a failure to evaluate threats after your opponent moved, whereas this involves a failure to evaluate threats after your own prospective move. Players who perform the first check but not the second are said to be playing "hope chess" because they simply hope their opponent doesn't have a devastating response to their move, instead of determining this through analysis.
(3) Analyzing lines, finding nothing good, then playing some other move without analyzing it, because you are tired. Don't rush moves. Don't play moves just because your opponent is online. A three-day turnaround will, in most cases, give plenty of time to perform analysis, do something else, and come back fresh, until you are satisfied. Use the timebank if you have to. (This assumes that you aren't playing too many simultaneous games. Since only strong players analyze quickly, you may wish to play a single or very few games at a time.)
Always make a move for a reason. Use your journal to list what you gain (and lose!) by every move. For example:
"1.e4 This move puts a pawn in the center, frees my king's bishop along the f1-a6 diagonal, allowing me to break a pin at e2, aggressively post my bishop on c4 pointing toward f7, or make a pin on b5; it also facilitates quick castling; additionally, it frees my queen along the d1-h5 diagonal, so that it can recapture on f3 (from where it eyes both f7 and b7) or check on h5, depending on the circumstances (queen/bishop/knight attack if my opponent isn't careful?)."
When your opponent moves, do the same thing for his move in your journal. Don't forget that his move may do more than one thing: the pawn move which obviously supports an adjacent, threatened pawn, may also allow free his bishop to make a withering pin on his next move, unless you act now to prevent it or develop a better threat to force him to meet you on your own terms.
Analysis: This is where you consider each candidate move, how your opponent can respond, and how you will continue, for each of his possible responses. Don't stop analysis too early. You want to reach a point in the game where the position is quiet, not just one where you manage to win a piece or pawn, then stop, assuming without reason that he can't win it back, or better. If you don't have time or energy for all this, do what you can regarding forcing lines and wait to see which move your opponent actually responds with: that will close off some possibilities, making it easier to look for forcing lines further along. I also like to explore to consider how a game might develop -- the kind of positions it might lead to, for strategic purposes, especially early in the opening or during quiet moments in the game.
One last thing: in a quiet position, strategy assumes greater importance; but during tactically active positions, analysis, not handwaving general principles is paramount. (Integrate strategy then, but don't use it in place of analysis when evaluating candidate moves.) Chess is concrete: consider tactical positions specifically.