If that happens, the change would be to tie budget of history and execution pool to the max size of block

I don't entirely agree with this. It might make sense to increase blocks in order to reduce short term rejection while not necessarily allowing longer term chain growth. In fact, if RC were robust enough (it currently isn't), blocks might even in some really theoretical abstract sense be unlimited as a matter of individual hard byte limit (an actual limit might depend more on execution time and other resources), but you wouldn't want long term growth to be unlimited.