An N95, in order to be N95, must filter out 95% of 2.5um particles. For larger particles, this is almost certainly an lower bound as well, but for smaller particles manufacturing techniques and tolerances can produce a wide variety of responses.
Since COVID has been detected in aerosolized droplets well under the micron size, as well as much larger particles in the 5-10um range, we really don't have a ton of information on how this translates to infectiousness for particulate sizes. We don't even have a good understanding of the degree to which infections can spread through tear ducts, which none of these masks address in any significant way. And from an epidemiological perspective, I don't think we'll ever find a signal; behavioral differences between regions is a far larger factor than mask type composition in a population.
That said, I think it is intuitively likely that higher filtration and better fitting masks will generally reduce transmission and infection, but I don't think we have an understanding of what those order of magnitudes are. Studies like this seem especially misleading. And some things are somewhat counter-intuitive; like the way that plastic valves actually seem to make things worse for liquid transmission because of their lack of absorbance even as they are more effective at filtering dry particles.