I may be wrong, but the way I understand it he couldn't do that even if he wanted to due to congress having ultimate authority to approve/disapprove spending?
It seems to have gone more in the other direction. At least in CS, if you want big funding nowadays you need to do DARPA research, because their budget has fared better. And DARPA research has gotten more micromanaged; gone are the days where they'd hand out large block grants to promote general American excellence in areas of "national importance". Nowadays they partner you with effectively a professional "minder" (called an "integrator") from a contractor like Lockheed or BBN, whose job is to make sure your research stays on the DARPA agenda and is going to deliver what they want.
Partly because I think we no longer feel much of a need to prove we're better in science/tech than a major rival, which was once the Soviets, so general science/tech advancement is a harder sell.
> Partly because I think we no longer feel much of a need to prove we're better in science/tech than a major rival, which was once the Soviets, so general science/tech advancement is a harder sell.
For example, reallocate defense spending to pure scientific research.