Take at look at
Gerards great PI vids and
Light Vortex tutes. There is much more in your data that you havent yet pulled out in PixInsight, you might be still using a simple workflow. Eg your background is too dark so you're losing dust lanes, there is more definition to bring out too plus star colours to bring back.
I'm wrestling with improving my own workflow and finding so much I can pull out without killing other parts of the image and those links above have some great information and techniques to help me understand and optimise PI processes. Something I've learnt is not every step will do something you can see in the images, its more about maximising signal and minimising noise so later steps can be really pushed.
I recommend you take your image through Gerard's tutorial for enhancing nebulosity. I'd bet you get something that will exceed all your PI attempts so far. its a very robust process that works well on most images. Some steps you might not need depending on image content , just think "structure/deep space object" instead of nebulosity, the process is just works great on faint fuzzies. The LVA tutorials have great gems of individual steps within the tutorials plus some real dogs of steps, so disregard the tutorial titles and just run through them with your own images to get a better understanding of what each step does and you take the parts you want and make your own workflow better as a result.
The links above are the only two resources I've found helpful in my PI processing, apologies to Warren's book and Harry's vids.
