The tl;dr version is that the link you gave, and what you appear to be talking about assumes that the "axis" is the one about which the disc is rotation, no matter what.
"Axis" and "rotation" are correlative, like parent and child, you can't have one without the other.
We're assuming the "on axis" is always perpendicular to the flight plate.
"We" are assuming nothing of the sort. The axis is the imaginary vertical line around which the disc rotates. You are claiming that a free spinning disc can have two at the same time, perpendicular to each other, which is akin to saying that single object can move in two directions at once.
So in my version, there can be torque components in different axes and in your version there can't.
Let us at least agree that torque is a force, energy applied rotationally. Torque from your hand to the disc stops the instant the disc leaves your hand. Torque, not being matter, can have no inertia or momentum. The disc, having mass, can have momentum. It can have rotational momentum, but because it cannot rotate on two axes at once, it cannot have rotational momentum in two separate directions.
Remember, the disc is spinning. The disc can have momentum, but if it is spinning on the yaw axis, any momentum on the roll axis
would have to spin on the yaw axis with the disc. What you are trying to claim is that the roll axis rotational momentum is somehow divorced from the yaw axis spin. In other words you are claiming that the "spinning system" or, possibly, the yaw axis itself has momentum. It doesn't. It is the plastic of the disc that has momentum. You cannot divorce the momentum from the matter of which the disc is made.
You won't apparently agree, but what you are claiming is that the momentum imparted on the disc by your hand changes direction when the disc rotates 180 degrees. Imagine an "x" marked on the disc directly under your thumb. If your thumb presses downward on the "x" as the disc is released, then the momentum of the "x" should continue moving it downward until the disc flips over. The problem is that the "x" is moving in a circle. In order to buy into your explanation, the downward momentum on that part of the flight plate would have to remain stationary while the "x" spins around the yaw axis.
Your explanation requires that the "momentum" be stationary relative to the yaw axis when the matter that it was supposedly imparted upon spins around that axis.
What really happens is wobble. The part of the disc that you "torque" downward goes downward, the opposite half circle of the disc goes upward, the disc continues to spin on the vertical axis and you see wobble.
Defining the axes differently doesn't change what's happening. The link you gave supports everything I'm saying. It's just that the context is different because lids act way different than disc golf drivers because of the difference in amount of mass near the rim.
It really doesn't support the claim that the disc has multiple spin axes. In fact, that page specifically describes wobble as a spin axis not perpendicular to the flight plate. It also clearly demonstrates a disc wobbling when it has a single, vertical spin axis, something that you have, at the very least, implied as being impossible.
That's just not true. Nearly everyone who's thrown with OAT has observed that happening and the link you gave described the same thing.
You are just begging the question as to what caused the roll. The roll is explainable without claiming that it was caused by roll axis momentum. But, just play with a disc a while. Toss it up in the air and catch it. Put a little "torque" on release and see if you get anything other than wobble. If "OAT" induced, roll-axis momentum makes thrown discs roll to the right all the time, you should have no problem inducing this when there are no other external forces to cause it (like the off-center lift combined with precession that causes turnover and fade).
You will get wobble after wobble after wobble. You will not get some slow roll in addition to spin. Post your video when you get the slow roll and I will buy into your explanation.
Incidentally, I have never heard a hyzer flip explained as being caused by "OAT." At least every once in a while, a XXX or NukeOS should flip up due to "OAT" when thrown on a hyzer, no?
With lids you get wobble real easy compared to disc golf drivers because of how different the mass is distributed in the disc. Lids require the cleanest throw in the tri-state area to be thrown with lots of power. With drivers there's a ton of mass at the rim (way more angular momentum) so they resist wobble really well. Players learn to throw these discs with lots of OAT (a lid would wobble a ton and crash almost immediately) to get them to fly straight. They only see the disc turn more than it should (but as much as the flight ratings describe) rather than the wobble you see with a lid.
They throw it with anhyzer. I've seen it many times.
I have also seen discs corkscrew on tomahawk throws. I suspect that an angle of attack that is below the nose down, zero-lift point is to blame, but I do not know that. This could also explain the inital turn on "torqued" throws rotating the spin axis too far forward just before release. It could also be a combination of an anhyzer release with and understable disc and low spin rate. Who knows. But "OAT"-induced, roll-axis momentum? Not.