To me, the central point of Matt's post isn't that deficits don't matter in a time of financial crisis and liquidity traps; the point is that, when Republicans aren't going to play ball no matter what, why not cram a bill full of things Democrats want? By refusing to vote for the stimulus package en masse, the Republicans have cut themselves out of the game. If some number of them would get on board, given the many large concessions that Democrats have made in hopes of enticing them, then they'd have something to bargain with. But by signalling that they were uninterested in compromise, they became an obstacle to work around or run over. If that's going to be the case either way, why not work to help the liberal cause?
Just to be clear, I don't think the Democrats need to be cautious about what they put into the stimulus in order to get Republican votes for it; I think they need to be cautious because when you embark on an enormously expensive gamble, you want to be sure that it's the gamble you really, really want to take. Given the choice, I'm pretty sure that most smart liberals would rather have a big-ticket health care reform - which is likewise a big political risk - than have the bill the House just passed. It's quite possible they'll be able to get both. But it's possible they won't - and the bigger (and more non-stimulative) the stimulus gets, the bigger a liability it will become for the Democrats if it isn't perceived as a success, and the more it will stand in the way, potentially, of the deeper reforms that liberals are hoping to attempt.
Chris Caldwell's analogy between the bill and the Iraq War is somewhat overstated, but I think he has the dynamic essentially correct:
The stimulus will be expensive, more expensive than the Iraq and Afghanistan wars combined and Nancy Pelosi, Senate majority leader, has called it a mere "down payment". The stimulus bill, whether it succeeds or fails, could be the Democrats' Iraq. Like Iraq, it is a long-standing partisan project that is being marketed as an ad hoc response to a national emergency. It reflects the pre-existing wishes of the party's most powerful interest groups more than the pre-existing wishes of the country. Democrats are now liable to be judged by the standard they created when they abandoned the Bush administration over the Iraq war: you break it, you own it.
As Caldwell suggests, the crucial issue here isn't how many Republican votes the stimulus package gets; the Iraq War got plenty of Democratic votes, and that didn't prevent it from becoming an albatross circling the Bush Administration's second term. The issue is the risk the Democrats are taking, period, by spending enormous sums that aren't obviously justified by the current crisis, at the start of an administration in which they're hoping to push through enormous structural reforms as well. Those reforms have something like a mandate, since Barack Obama campaigned, and won, on promises to fix the nation's health care system and reform the energy sector. But he didn't campaign promising to spend massively on existing government programs - and the more massive the forthcoming burst of spending gets, the bigger the risk that it will end up swallowing the broader ambitions of this administration's first four years. This doesn't mean liberals are wrong to take the stimulus money and run with it, but they should at least be clear-eyed about the political risks involved.