There's a page eplaining CMB physics here, which might be useful to people trying to understand the polarization stuff:
http://background.uchicago.edu/~whu/

It does get quite technical though. The scattering of light from free electrons does depend on polarization. The reason you end up with a net polarization at some point rather than everything cancelling out is because radiation reflected from different directions ends up with a different polarization. A region of the Universe will receive different amounts of radiation from different directions (because the Universe isn't completely homogeneous).

The signal is weak though: polarization measurements are much harder to do than total intensity measurements.

Regarding the amount of 'ordinary matter' in the Universe, current estimates are that it makes up about 4.6% of the critical density. Dark matter makes up about 22.8%, so you end up with a total of about 27% of the critical density coming from matter. The rest comes mainly from dark energy, with a very small contribution from radiation. It's thought that the total comes to something very close to 100% of the critical density.