1. I don't think so. I think this is because the "moments" are different. A star becomes unobservable once it passes from the region within the particle horizon

1. I don't think so. I think this is because the "moments" are different. A star becomes unobservable once it passes from the region within the particle horizon to the region outside the particle horizon. But passing into this region can't affect all the information on the way to us. That's because the region doesn't exist until we trace all light path backward. So start right now and trace out 14B years of light paths. You will find that star is within the region. However some point in the future you notice that the star passed beyond the horizon. The star still exists it still emits light, but that light will never reach us. We don't see the star "go away" as it passes beyond the horizon. Instead one day the future we will receive the last piece of information about the star redshifted to infinity. I think this means we observed a decay to some constant and that constant should be zero. This argument feels very similar to black hole arguments, so rings true to me. 2. I think this is trying to state something paradoxical. But it's not really a paradox. Everything we observe is within our particle horizon. Everything has it's own particle horizon. These regions don't overlap. I think if I restate what you said without the objects the paradox dissolves. > If there are points (P) in our particle horizon far enough away from us that these newly unobservable (U) points are still within the respective particle horizon of the points (P) in our horizon, should we detect the points (P) our particle horizon as being affected by points in their own horizons? Of course, why should what we can observe in future affect what had happened in the past at some distant point.