1. fun isn't it?
2. do you have a logic analyser to watch what is happening?
4. try to find a flash drive with any of the following controllers, as these shouldn't have XOR...
UT165-L48, UP10, SM321QF, SM3251 Q BA, SM3252 Q BB, SM3253 L AE
5. maybe and maybe not. the fingerprint, if a digital representation is stored on the flash, would be VERY interesting to analyse. but it may be just an unlocker to the NAND, and controller-to-NAND is same as any other. I may have to look into these. Maybe take it apart and see how many chips. If there is an extra eeprom for example(could they be this stupid?) then probably nand is normal. One thing to check is the marketing material for the flash drive. If it mentions encryption, it could be a waste of time..

3- All I want now is to find a way to understand how to do the pinouts correctly when I don't have acess to the datasheets.

Oh wait when you get to Monolithic

3- All I want now is to find a way to understand how to do the pinouts correctly when I don't have acess to the datasheets.

It would also be helpfull to understand why I'm only being able to read for example 4 MiB out of 2GB flash drive ( i assume 99% it's bad wiring but I just can't guess :x)

ISTM that the most logical way to analyse the wear levelling and flash translation methods would be to write a visually recognisable data pattern to each LBA. For example ...

LBA (h) xxxxxxxx
(dec) yyyyyyyyyy
pass #nnn

You could write your own tool to do this, or you could create an appropriate 4GB file and then clone it to your flash drive using Linux dd or a disc editor. You would increment the pass count for successive clones. Spildit and I have used this method to analyse the cache RAM structure of a Hitachi HDD. You could also use such an approach to analyse a RAID.