Yeeeeessss..... holy cow this took me forever to fix... the funny part is that i'm not even sure what did the trick X-D
I just realized that the board was giving me a very random cart check in the test, sometimes the cart was found, sometimes was not... and even the sram fill was very random.
First i changed the 62256 next to the bios, my programmer tested it as good but i've changed it anyway.
Then i went to check the hc245 buffers near the resistor arrays, they basically make a bridge from the cart to sram/Vrender. They were all tested good, however when i tryed again after put them back in the motherboard the game started to boot all the time but it was soft-locking after the character selection screen.
There was no rust near them or underneath (the board is in superb conditions) so i was suprised on what the hell actually happened doing so.
When i started to work on this board months ago i changed the 3.3v regulator, using a buck DC-DC converter (the original LM was rusted all the way), but after checking carefully my power i realized that the output voltage was 3.05/3.1 and it was probably not enought.
I then tryed with a small solution, just to save some time, and i used an AMS117 3.3 in sop. Game started to working better, however it did not past 10/15minutes of working and it keept freeze again. Was probably not good for the board amperage and i did not keeped attention to the datasheet.
Then i just bought an LM317 with ADJ, arrived today, and used 220R and 390R to obtain 3.4V and nearish 1.5Amp. After that, it worked straight for an hour, i turned it off and consider it as work done.