Another explanation could be that the money givers haven't gotten as much health improvements as they would want because they haven't directly asked for them, as this paper acknowledges:
"A recent focus on productivity has dominated the competitive evaluation of scientists, creating incentives to maximize publication numbers, citation counts, and publications in high-impact journals. [...] The efficiency of society’s investments in biomedical research, in terms of improved health outcomes, has not been studied."
Computer programmers do not produce more profitable programs if you pay programmers by lines of code, either.
And yeah, I heard Casadevall speak at a conference and "used up low hanging fruit" was one of the theories. It's less likely that there'll be another penicillin moment.
It's all pretty depressing, and very true that it's all driven by Science/Nature/Cell publications, not by discoveries.
Increasing disparities between resource inputs and outcomes, as measured by certain health deliverables, in biomedical research http://www.pnas.org/content/112/36/11335.full
Basically, increased funding of biomedical research hasn't resulted in better output, and that we do, indeed, need to change the model. The current model is really inefficient.