At this point I think the problem is price. You only see white collar workers on the express because a round trip is $200+. If any of these people (myself included) had to foot the bill for the train tickets I doubt the trains would be nearly as full as they are.
Point being: the Acela (IMHO) doesn't qualify as a high-speed train if only because it doesn't "feel" high-speed compared to the other options. Commuting in 3.5 or 4 hours is emotionally irrelevant. Commuting in 1.5 hours would feel remarkably different.
1) Only chosen to buy 20 Acela trainsets, with one first class, one cafe, and four business class cars each, and I believe four of those are always scheduled to be undergoing maintenance at any given time.
2) Told Amtrak to try to make a profit
Given the constrains of needing to somehow decide which people get to use the inadequate number of seats, and trying to make money, how do you kill two birds with one stone?
The other thing is that greater superelevation would increase speeds possible through the existing curves. But as you increase superelevation, you potentially increase the minimum safe speed through that segment of track, and having 0 MPH be a safe speed is desireable. There may be some maglev derivative that could help with that problem. Rubber tires on concrete also have greater friction, which may allow them to tolerate a greater amount of unbalanced superelevation.
The really interesting thing about a 90 minute or better Boston to New York City trip is that it opens up the possibility of daily commuting. It would also have the potential to add places like Springfield and Hartford to the set of suburbs of Boston and Manhattan. But Amtrak seems content to plan on that not happening for another 30 years.