I like to know what the response is and I don't always believe what I read. SPRAT is a nice journal but who knows whether what they print really does what they say. Filters are pretty arcane devices. I use Amateur Radio Designer to look at all filters I intend to use in a homebrew project.
In the case of the 80 meter preselector filter, I don't doubt it does most of what they say. I question the bandwidth. I get a much sharper peak than you specify. So much that attenuation at, say, 3.5 mHz is much larger than at 3.6 mHz. I think the filter should be tunable! As far as the receiver response is concerned, I have found in the past that if you hear anything it will immediately be 100 times better than any receiver you buy! It's magic. All measurements go out the window when you hear the first CW signals! I built a receiver back around 1979. It was a DC receiver for 40 meters. I included several circuit designs to improve the sensitivity an selectivity of the unit. It worked very well. In fact I thought it was as good as my TS520. Then I made some measurements of sensitivity. It measured 2 uV for a 10 dB signal/signal + noise sensitivity where my TS520 was under 1 uV! Ears can be receiving!
I would suggest you do a bit more measuring and optimization now and then add the new stuff!
By tuning I meant a front panel control. The filter is not broadbanded enough. If you presented a fixed strength signal to the receiver, a signal generator for example, you would notice a big difference from band edge to band edge!
Building is an ideal way to learn. You see your results right there on
the bench, but you should make some measurements to really understand
what you have!