"Newton had already shown how mass can be treated as a point location (center of gravity), which eliminates all density issues. This, I'm confident, is held to be true in GR."
Maybe I should be starting a new thread, but this seems like a good thread to enter on this specific point.
With respect to JZZ's query, I have some input from work I've done involving orbiting tether structures. An orbiting tether rotating on the orbital plane will have a natural orbit slightly lower than an equivalent point mass - this is because of the inverse square relationship. If you wanted to deploy a tether (with symmetric mass) that orbits in a vertical tidally locked position, a little boost is required to set it on such an orbit along with the generation and release of the necessary angular momentum to produce the desired rotational frequency. For the internal rotation of Mercury, I don't think it makes a difference at the Newtonian level because the rotation doesn't create a shifting centre of attraction like a rotating tether (except tidally locked rotation).
With respect to Helio's point quoted above, I don't believe Newton, I think he has this wrong and I've made a simple tabletop experiment that I believe proves this. This is what I'm here to share:
Getting to the brass tacks, Shell Theorem tells us that spheres attract each other through their centers of mass. Shell Theorem doesn't equate spheres of equivalent mass and different densities/diameters despite being claimed that it does. Throughout the Shell Theorem operation, the components of gravity perpendicular to the attraction vector aren't included, they're cancelled out. These gravitational components, however, are operating and become more significant the larger the sphere is relative to the proximity of the attractor (the point outside the sphere from where all the calculations are based from).
If we consider the gravitational effect of our own bodies on continents remote to us, a simple force diagram shows that our bodies have the effect of pulling the remote continents together. This can be simplified to two force vector components - one towards the centre of mass of the two continents being measured and one orthogonal which is experienced as gravitational compression between the continents. This compressive force is counterbalanced by one of tension active on our own body. Simply, our presence pulls the continents together and the presence of the continents pulls us apart. If somehow the Earth could be magically shrunk, the effect of the same amount of mass would be causing increased attraction and less of the compressive and tensile forces. It's just force triangles, nothing complicated.
Was Newton's objective through Shell Theorem or any of his other work directed towards ascertaining big G? I don't know that it was, I think his intention was the curiosity of understanding the motions of heavenly bodies matched with a strategy to calculate the center of attraction.
I have experience posting ideas along these lines on other science forums when they were less solid, but I've now figured this out in a top to bottom kind of way. I've never received a solid criticism which has led me to believe otherwise and it's come before some serious people, not just in science forums. I don't think this invalidates GR because if correct, we just have a calibration error that doesn't affect GR as a principled theory.
I will gladly accept evidence to the contrary, but I believe big G is slightly yet considerably understated.
I have lots to discuss and lots of diagrams to assist explaining my body of work but I'll do that on another thread assuming this post is not called out as an obvious blunder.