I think this is a pretty good argument for why the basic keynesian model isn't the whole story, but that isn't a reason to dismiss the whole model as a positive description of how economies work. It has several basic predictions. On the monetary side, increasing interest rates tends to reduce output and there is a positive correlation in the short run between inflation and output. These are two of the strongest empirical facts about Macroeconomic data in the last several hundred years.
Of course, everyone in academia realizes that these models aren't enough to model the crisis. I think an honest economist may tell you that the problem with the models is that they assume people are too smart and predict the future too well. But the solution to that is not to throw everything out the window, but model beliefs and preferences of consumers better. A lot of the more promising work is about the idea of a leverage cycle and the corresponding asset booms. But of course you need stupid consumers to get a leverage cycle.
My bet is if the Austrians actually wrote down their models, they would find the same type of irrationality underlying a lot of the results.
Another approach economists are taking in response to the crisis is to more explicitly model the financial sector of the economy and what we call financial frictions that prevent economies from achieving optimal outcomes.
