The two computer screens depicting completely different subjects have almost everything in common, in that they are of the same material. However, where they differ -- namely, the color of each pixel -- is where all the information about the painting is contained. So the screens have enough different information (at the quark level) to distinguish what the paintings are about.

But the two screens with two different subjects are probably more similar than a screen and a painting with the same subject, in terms of coordinates of quarks. Additionally, it's not clear to me that there's a one-to-one correspondence between color and quarks. Even establishing a correspondence between color and chemical make up is extremely difficult, due to the influence of natural selection in how we see color (I remember Dennett having a cool chapter on this in CE.)

I don't want to make our disagreement sound more stark than it actually is. I agree that the about-ness is in the mind of the beholder, and the stick figure is a good example as well... but I think this just emphasizes my point. Let me put it this way: Given the data for the point-coordinates of the three entities, could a mind choose which one had which subject? No, even though the criteria is buried abstrusely somewhere in there. The point being that the models are inextricably separate in the imagination, and its therefore not clear to me why its a priori logically necessary that they all collapse into the same territory (though I agree that they do, ultimately).