I would first say that I agree with the reference you cited that there isn't any observation that definitively settles the question - as far as I know. (I haven't attempted to check other references on the point, but the point seems valid to me.)
However, I would also say that I couldn't produce a theory in which the Sun orbits the Earth that's valid for all of space-time - at least not easily. The details and underlying issues are a bit technical, basically though locally there exists a fermi normal frame in which the Earth as it rest, but this frame is only valid locally, if you try to extend it to far it, one runs into mathematical difficulties.
So while it's true that GR allows one to determine a "local viewpoint" for any observer, in which they are at rest, you need to read the fine print about the view only being local. If you want your theory to cover all of space-time, it's harder.
I can't prove, at the moment, that it's utterly impossible to find a theory covering all of space-time with some "relaxed" notion of a viewpoint (i.e. perhaps a viewpoint is not NECESSARILY equal to a fermi-walker coordinate system), so I'll make a lesser claim. I don't think you'll find a _published theory_ that starts out assuming the Sun orbits the Earth.
Furthermore, saying that you do sounds a bit - odd. So I'd avoid saying it.
If you consider Newtonian theory, it will tell you that the Earth and Sun orbit a common barycenter.
If you consider general relativity, it will not single out such a point directly. Instead it will say something along the lines of ""Mass tells space-time how to curve, and space-time tells mass how to move."" (a quote from Wheeler,
http://astronomyandspace.blogspot.com/2010/06/quote-john-wheeler.html).
In GR, the idea of an orbit is an approximation anyway. It's a pretty good approximation, but you'll see things like "orbits decaying" that don't have any Newtonian counterpart. And if you try to take your thesis really seriously, you'd have to settle issues like "what is the center of a decaying orbit".
So, I suppose, my best answer is that the basis of GR is not really built around orbits, orbits come about as a good approximation but aren't a fundamental part of the theory - orbits come about when the system exhibits repetitive, or nearly-repetitive, motion.