Regarding the aim point for the gun sight.
The collimated gun sight uses a lens to collimate the emerging image of the reticle such that said reticle appears to lie at optical infinity. Or at least at some considerable distance. This is to eliminate parallax as the observer's eye point translates laterally.
In other words, if you have the plane lined up on a distant target. No matter how much your eye moves about the reticle will remain in place on that target. It should not move to any meaningful degree at all. The reticle in effect remains 'locked on' the target as long as the plane's aim is unchanged.
In our game, the usual practice is to locate the reticle roughly 10m ahead of the gun sight or pilot. This is far enough to reduce parallax to a small extent if your sight line passes through an edge of the reflector glass (as opposed to through the center.) As I noted, in reality the reticle appears to lie hundreds of meters (or farther) away.
In the short video Andrew posted, it looks to me like the reticle is hardly any distance out in front of the reflector plate. This reflects a common misunderstanding that the reticle's image is projected *onto* the glass, as though said glass plate were a screen. Either that or that the reticle image is but a very short distance ahead of the reflector, at a distance about equal to the space between collimating lens and angled plate above it. Either instance is most decidedly not how it works. The reflector is not a 'movie screen', of sorts, but is instead a simple 'window' that acts like a beam splitter.
If one were to peer straight down into a gun sight, directly into its collimating lens, one would see the reticle in the same way as in the usual mode through the reflector plate. It would appear to lie at optical infinity, far beyond the actual, physical bottom of the sight.
A good way to verify this fact of collimation to infinity is to photograph the reticle through the reflector, as seen against a reasonably distant scene beyond. If the camera lens aperture is set wide open, where the depth of focus is quite shallow, when focusing on the nearby reflector plate the reticle will be well out of focus. Conversely, set focus on the reticle and the reflector plate will now be out of focus. If the reticle image were focused upon the glass plate (or hardly any distance out the n front of it), both would be in or nearly in focus simultaneously. Indeed, when the reticle is in good focus, so is the distant scenery beyond. The reticle's best focus will occur when the camera's lens is set to infinity focus, or at least near enough to it.
Indeed, in many online photos and videos of working sights, the reticle and reflector plate/sight are typically never in good focus at the same time. And the reticle is in good focus along with the distant scene it is projected upon.
By way of bona fides, note that I've worked professionally in optics. And I've built collimated sights for distant target alignment, as well other optical instruments.
I'm going to download this fine 'pit. If the reticle needs a more distant placement, such will be effected and made available. And my own slightly more accurately modeled PBP sight will be fitted.