"What is true is that there is most definitely still is a 'mainstream' set of assumptions that are somewhat difficult to challenge and which translate into economists are conditioned as to 'what to think'. Try for instance to present a seminar where agents in your model have and exercise any form of market power and you will be immediately and repeatedly challenged (as the Card discussion above suggests). Economic rents are just not supposed to survive, and your audience will grill you until you surrender (if not in this seminar, in the next paper you choose to write).
But present the same seminar and begin with a likely even more improbable statement such as 'I will assume free entry, competition and zero profits in every sector' and it will probably slide by without challenge."
One could continue to question why the heretical challenges brought up by heterodox economists are quickly dismissed, while obvious impossibilities (boundless rationality) can be so easily accepted.