It is not really a secret, I simply don't have a standardized way of processing. This makes it difficult/impossible to explain what exactly I did, because it was a lot.
Sharpening of the images was entirely done in Photoshop using a mixture of techniques, mostly done by just experimenting a lot. I know I did unsharp masking, and some modest smart sharpening, but I also emphasised a little bit the darker contrasts in the image.
But perhaps the most important thing is that the image is an R-RGB image. I actually combined two R recordings, and added a little bit of green as well (20% or so), to create an artificial luminance channel. This made for relatively noise free data to work with.
I have also been experimenting with several different noise reduction plugins lately, I don't like the standard Photoshop noise reduction.
Basically I use a lot of tools, but pretty much everything with moderation. Too much sharpening or too much noise reduction gives you results I don't like. If you think an image is sharp enough, you have probably sharpened it a bit too far already.
But part of the fun - at least for me - is to simply experiment a lot. If I could get a perfect image with just a single press of a button, most of the fun would be gone.
But perhaps that is different if you can image every night