You've given me tons and tons of data but only a little bit of analysis. Push yourself to analyze this more thoroughly - that's how you're really going to learn.
I'll respond now because I know test day is coming up for you, but I actually do recommend that you stop reading right now and go think about this data more yourself - and what you think it means, and what you should do as a result - before you keep reading. You're going to need to be able to "read" this data all the time and react accordingly, so learn how to do so right now!
NP at 50% but also really fast? That might be an opportunity to get better! (Look and see what those problems were though - maybe you just got a few killer combinatorics and probability - and those aren't worth blowing a bunch of time, especially when they're really hard.)
FDP at 36% - yes, need work there. Are these scattered across all categories? Certain ones? Primarily word / story problems or more pure math?
Major conclusions (you can match with your specific category areas):
1) Low-hanging fruit = careless errors. You can already do these! Figure out why, figure out what good habit to make or bad habit to break, and get those points next time! (Some of these will be due simply to speed / timing issues - fix that timing and these will go down.)
2) Low-hanging fruit on the other end = categories that are consistently slow and wrong (eg, sets, probability, ...). Get them wrong faster; spend that time elsewhere. In other words, no, don't spend time on category 5. :)
3) Category 4: these will need to get split into two: the ones you can get more efficiently and the ones you can't. For the ones you can't, if you really are losing 45-60+ seconds (above the average) and you still have careless errors on other questions, get the too-hard ones wrong faster and pick up those careless error points. (The longer you spend on something, the more likely you are to make a mistake - so I'd rather have you answer correctly the stuff that you can do more quickly.)
Categories 1 and 2 are your "good" categories - yes, it's actually good to get something wrong in the appropriate timeframe (category 2). That means you did try it, but then you couldn't do it and so you moved on. (Of course, as always, keep an eye out for careless errors even in this category.) I'm not too concerned about that as long as you've got other low-hanging fruit to deal with first.
Now, of course, you can always go back to category 2 and try to get better longer-term... but in your case we're only looking at a couple of weeks anyway, right? So you probably won't have time. Either way, always deal with the lowest-hanging fruit first.
A few resources to help based on the detail you provided (but I'm not giving you much because, again, you don't have a ton of time!):
translations / dealing with story problems (can help with algebraic translations, FDP story problems, any word problems in general):
http://www.manhattangmat.com/blog/index ... into-Math/NP:
http://www.manhattangmat.com/blog/index ... -gmatprep/Verbal
CR Weaken (just in case this helps you shave time):
http://www.manhattangmat.com/blog/index ... g-problem/http://www.manhattangmat.com/blog/index ... n-problem/So do two things:
(1) low-hanging fruit on weaknesses
(2) comprehensive review, with a little more focus on strengths and medium areas - make sure that you really can get all those points in category 1 and that you're not forgetting anything because you haven't looked at it in a while (because you weren't having problems with it).