no,... it is NOT "ALL" polarizations when fed 90 out and Xpolarization.
this is basic turnstile antenna theory.
when 2 antennas are @ 90 degrees to each other and are fed 90 degrees out of phase and mounted horizontally the antenna is nearly omnidirectional on the horizontal plane.
When mounted vertically the antenna is directional to a right angle to its plane and is circularly polarized.
these antennas are normally used when working the FM sats, but, they can be used on HF with some success, go here: Circular Polarization on HF
We will have to disagree on this one then. Try various angles with a dipole when receiving a circular polarized FM broadcast signal. As long as the dipole remains broadside to the received signal, you can move the antenna from vertical to horizontal and not see any change in signal at any angle. There is not a peak at the vertical and horizontal points when the signal is truly CP.
You'll notice no where in the links you give does it suggest that a CP signal is only polarized at 90 and 180 degrees. In fact it explains that this system sets up a rotating electromagnetic field. This rotating field is present in all polarizations in TX and remains responsive to all polarities in RX. Depending on the orientation of the two elements, it's also possible to have a nearly omni directional CP signal. It's done every day on the FM band in designs like the ERI Rototiller.