Even if this is a runner logically, we are still faced with two challenges, to motivate it and to identify the levels of different beliefs.

Motivation might come from the fact that we are talking about belief. Beliefs are different from physical things, including the states of brain cells which correspond (on a token-token basis) to beliefs. Specifically, we have intentionality, which opens the door to self-reference, the problem which in a different guise created the need for the original type theory. This is only a sketchy motivation, but it might do for starters.

Assigning levels to beliefs would be easy a lot of the time, but there would also be hard cases.

Many beliefs, certainly the vast majority of beliefs held by non-philosophers, are clearly first-order. A belief that epistemologists discuss and disagree over principles of equal weight is also first-order. But the definition of the class of first-order beliefs would not be easy.

EW1, EW2, and so on are clearly second, third and so on order. But what about other epistemological principles, for example that you should believe the logical consequences of your beliefs, or Bayes’ theorem? Their order might depend on the order of that to which they were applied, so they would need to be split into principles of different types too.

If we got all this to work, where would it leave EW? I could accept (or reject) EW1 without further reflection, but that would be reprehensible. I should think about whether to accept EW1. Knowing that there was disagreement about EW1, I would need to decide whether I accepted EW2 in order to reach a verdict on EW1, and so on up the chain. We might eventually reach an EWn on which there was universal agreement, but I would be surprised if we did. It is more likely that we would give up the pursuit and join Mr Hume for dinner and a game of backgammon (Treatise 1.4.7, para 9).

Finally, Russell perhaps wisely went no further than this down the road to EW:

The skepticism that I advocate amounts only to this: 1) that when experts are agreed, the opposite opinion cannot be held to be certain; 2) that when they are not agreed, no opinion can be regarded as certain by a non-expert; 3) that when they all hold that no sufficient grounds for a positive opinion exists, the ordinary man would do well to suspend his judgments. These propositions seem mild, yet, if accepted they would revolutionize human life.

(Skeptical Essays. I found this on a website which did not give a precise reference, and I have no copy to hand so I have not checked the text.)

http://www.oxfordscholarship.com/oso/private/content/philosophy/9780195131932/p108.html

For those of you who don’t have a subscription, the salient section is called “pluralism”, and it starts on page 437.

http://www.veritas.org/media/presenters/201

The talk is called “Pluralism: A Defense of Religious Exclusivism”.

There (back in 1996) he charges that Hick’s position is self-referentially inconsistent.

The exchange between Hick, Plantinga, and Van Inwagen is a foundation of the current discussion of peer disagreement. (I think that Plantinga and Van Inwagen did “gazump” you in print.)

Here’s the reference for Hick’s article:
John Hick, “The Epistemological Challenge of Religious Pluralism,” Faith and Philosophy (1997: 14:3), 277-285

Plantinga’s article immediately follows, and Van Inwagen’s follows Plantinga’s.