But given the scope of pfSense 3.0 plans, it's hard to see how that could get done in a reasonable time frame without more people at ESF, or greatly increased community contributions, or both.

Also, I'm not sure I see the logic in removing PHP code from the 2.x line of pfSense.  Yes, I can see that as a goal for 3.0 and I can see maintaining a 2.x branch for slow movers, but you lose all of the stability of the 2.x branch when you do a massive re-write like that.  I see stability as one of the prime reasons to keep 2.x branch going, not just to accommodate 32bit users.  Unless I'm missing something.