Yes, I meant unstretched.

No, I didn't mean all matter-constituents expand, only the distance between non-bound matter-constituents (i.e. the gaps in between the galactic clusters).

In the standard model (Lambda-CDM) we theorise that the observable universe was only around 40 million light years in radius when the CMBR was emitted, and what was the edge of the observable universe then is known as the "surface of last scattering" and is now considered to be around 46 billion light years away. Our observable universe is now around 1090 times the size it was when the CMBR was emitted, and the surface of last scattering is estimated to have a redshift of around z=1089.

The z factor equates to the difference in the scale factor of the background metric between emission and observation, so when we look at a galaxy with a redshift of z=1 we are looking back to a time when the observable universe was half the size it is currently. To put it another way, the scale factor is usually written as 1+z, so the universe is now twice the size it was when the light was emitted from a galaxy with a redshift of z=1, is now 3 times the size it was when the light was emitted from a galaxy with a redshift of z=2, is now 4 times the size it was at redshift z=3 and so on... so I shouldn't really have said "seven times larger" for z=7, it should have been 8 times the size, and it is easy to get a little confused with the semantics here! (This is why people don't like "word salads"). But that is what the "z" in cosmological redshifts represents, in a nutshell.

So if we imagine the light leaving a galaxy that we measure as having a redshift of z=7, when that light passed a galaxy with a redshift we measure as z=6, the universe had doubled in size since that light was emitted. As it passed a galaxy at z=5, the universe was 3 times the size it was when that light was emitted. So as it passed what we measure as z=4 it had grown to 4 times the size, at z=3 it was 5 times the size, at z=2 it was six times the size and as that light passed a galaxy we measure at z=1 the universe was then 7 times the size it was when that light was emitted. As it reaches us (we are at z=0!), we find the universe is 8 times the size (1+z) it was when the light was emitted.

You are correct to ask why light should expand with the cosmic expansion, and I have no answer except to say that however we conceptualise the mechanism behind it, we measure the wavelength of light as having been stretched by the same factor that we think the universe has expanded by since that light was emitted.