Ok, now that I can see the board with the chip removed, I am spotting a few potential trouble spots for breaks. Where the outer pad is no longer connected to the inner pad and could have been the source of your problems. When soldering in the new chip, you added fresh solder from the tip of the pad and it flowed inward, bridging the breaks. (potentially).
Here is a diagram explaining what I mean: I marked in read the trouble areas, and marked in green the breaks
YAY, serendipity rocks! Found info on this during a ridiculous 24 page sifting expedition in Hacks & Development. Sounds like mouse function would be the impediment:
I'm wondering if the USB Wombat from @bigmessowires can bring mouse function back up to snuff at 20MHz? That's the target clock for my KLUGE (insane functional equivalent) I'd almost, but not quite wager its ADB input is a far cleaner signal than a 30 year old mouse can achieve? Thinking clock synchronization/bus speed borkness?