Yeah, I agree.
However I'd suggest another round of balancing and 3 months of release from that result
before a board reset to make sure, as mistakes were made by the playtesters (of which I was one, I'll add) about just how powerful some of the changes are.
Assuming everything is more balanced (doesn't have to be perfect) after the 3 months release phase, then a board reset would be a good idea. If it's not then another month of internal testing + 3 months of release would be a good idea.
It might seem long-winded, but testing game mechanics properly always is.
Currently Vegas is over-powered. It should be able to get high scores based on luck, but not to the degree it currently can.
In short it should require as many plays to get a luck-based high score as the other characters require to get a skill-based high score. That varies per player and I suspect someone else has a better idea for exactly how to do it (indeed we have threads full of them).
Eraser is slightly over-powered, but not to the degree Vegas is. It needs a little nerf, not a big one.
The current problem is that nerfing any character would require a full board reset. During a beta test phase this wouldn't be a problem, but for the confirmation it would. A simple solution would be to increase the score of the blocks by 1 across all characters for the confirmation phase. There's probably others, I've not thought too deeply about.

I don't think the others need changing from what I've seen on the boards, but I only tend to play Mono and DV, so I'm not the best source on that.

I think overall this is a good idea, but I suspect they'll be a lot of people who complain about it. There always are on board resets, unfortunately.
An alternative to a full board reset might be to keep the +1 boosts of each block used during the release phase, which should see the old scores slowly filter out, although I do think some of the more ridiculous Vegas scores would still need removing.