1sataker wrote:
For LR, I hightlight the conclusion and premise if there is an argument and write down the explanations I thought important and would like to remember.
And review why wrong AC is wrong and why correct one is correct.
If the explanations are too long, I print them out and highlight.
This is great for understanding the question, but not so great for understanding your reasoning and how you're going to not make that mistake again. Add these questions into your review:
1) What about the right answer made me eliminate it? This will help you avoid falling for a trap that they use to hide the answer.
2) What about the wrong answer made me select it? This will help you avoid falling for a similar trap.
3) Next time I see ______________, I will __________________. This is the most important one - what are you going to proactively do to avoid making this mistake again?
Most people, when reviewing, just want to understand the test, not their own faulty reasoning. But you can't get better solely by understanding the test - you need to also adapt your thinking to it.
For RC, I hightlight the parts that gives the reasons why correct answers are correct in different color, and review the answer choices I was not sure. Again, write down the explanations if I think important.
Definitely cross-check the correct answers with the passage. See where they're drawing the information from. There's a pattern to the types of statements they ask about.
Also, when reviewing, make sure you analyze the language of the answer choices. You can usually get rid of 3 ACs without even reading the passage because they're too extreme to be something an author would commit to. If you start paying attention to language, and add it as a way to eliminate answers on top of content reasons, you'll be in good shape for RC.
For LG, I watch videos especially for the ones I got wrong, and if have time re-do the questions I got wrong.
You need to tie the methods that you see in the videos to the game type so that you know how to approach a similar game in the future.
Also, the LSAT has, recently, been taking elements traditionally associated with one game type and throwing it into another. The hallmark example of this is the mauve dinos game - it's an In/Out game, but it has characteristics, which is traditionally an ordering twist. While reviewing a game, think of it on two levels - what basic game is it like (ordering, In/Out, grouping, hybrid), and what twists are on top of it. There are tools for the game types, and there are tools for the twists, and developing each toolset independently is important to having the flexibility needed for the modern exam.
But I feel-and actually it is-really time consuming...and I'm not sure whether those work will pay off.
Also not sure how those work help me to learn LSAT...
Should I stop doing those work? Am I just wasting time?

How do you review&what is the correct way to review?
It is absolutely not a waste of time, and it is absolutely going to pay off. There's too much bad advice out there that says, "Just keep doing problems until you do well." Instead, it's important to spend time properly reviewing questions so that you actually improve. Every question that's even been on the LSAT has shown up many times, with just small details moved around (for example, every Strengthen question with Causality is, at its heart, the same question). If you can really understand it by reviewing one you got wrong, they shouldn't be able to trick you on that question type/answer type combo again.
Doing more questions is obviously "better" than not, but you might find that, with proper review, you don't have to do 75 PTs worth of questions to get to where you want to be on the exam.