Yes, system improvements are addictive!

A lot of newcomers to the GHz bands ask me about how to improve the performance of their systems, andI reply that it's down to transmitted ERP and receiver noise figure. While QRO is king, noise figure is an altogether more subtle thing. Improving noise figure does not quite give you the same improvements when your dish is looking at the "hot" horizon as it does when looking at cold sky for EME. The "bible" on this is (The sadly recently deceased) VK3UM's article so read it, I don't intend to repackage it here! Receiver performance depends a lot upon on how much coax/waveguide you have between the input of the preamp/transverter and the dish feed. Remember that every 0.1 of a dB loss between the feed and the front end adds 0.1dB to your noise figure, but a 0.1dB improvement in Noise figure can give you more than  0.1dB improvement in signal to noise under certain circumstances. So your first aim should be to reduce this front end loss as much as possible by putting the preamp/transverter close to the feed. You can get a real feel for all this by downloading  VK3UM's) "EMEcalc" and doing some "what ifs" with the RX performance page. There you can change gains, noise figures and antenna temperatures and see how it effects your output S/N ratio.
In my case, my 10GHz transverter/PA is in the loft by the gable end and I have a run of about 5m of FSJ2-50 up to the dish, giving me around 2dB loss. The dish mounted 1dB preamp makes a notable difference, as I can switch it in and out of circuit from the shack.
Dish upgrades are always a good thing, giving you better RX and more TX ERP but remember the narrower beam width will make it all that more difficult to point. I find my 60cm dish is OK with a bog standard G450 rotator on 10GHz but it is a pain on 24GHz. Also make sure that you get the elevation spot on (more difficult) I cheated and used an elevation rotator as well but it's all weight on the pole. EMEcalc allows you to calculate your dish beam width.

See my August 2015 "GHz Bands" column in RadCom for more ramblings about preamps and noise figure.