This is a great critical analysis of Van Fraassen's ideas, Matt -- I really enjoyed reading it.

You say in conclusion Van Fraassen "makes several assumptions that invalidate his position." The one you cite in that paragraph is "one can be justified in believing in observable entities."

In an Occam's razor exercise, I'm tempted to throw out the assumption you cite, then agree with Van Fraassen, then reject "philosophically controversial claims" and simply let science make empirical progress as it has been doing.

The best way to answer the question "what is real?" is to say "nothing is." "Reality" is a concept empiricists can do entirely without. The minute you introduce "reality" into the discussion, you have a term you really can't define -- then end up trying to square the circle or cut Solomon's baby in two.

What does exist is phenomena. The early empiricist George Berkeley made an excellent case for phenomena and against "reality." He said God makes phenomena, that reality is unnecessary as an explanation of phenomena.

Three centuries later, we see that God is not necessary either -- we just plain don't know the source of phenomena. In fact, we don't know if phenomena even have a source. I believe there is a source, but that it is entirely "unobservable" -- the source is unseeable (as explained further in some of my truth units pieces on Medium).

An argument similar to Van Fraassens' is made in a book I read: "The Blind Spot: Why Science Cannot Ignore Human Experience" by Adam Frank, Marcelo Gleiser & Evan Thompson. You might find it makes a better case than Van Fraassen. It grounds science in human experience, rather than observables.