I can offer some more thoughts about the front end design of the radio if this is of any interest.
If you look beyond the first RF amplifier and BPF coils you arrive at the first mixer.
This is usually an active mixer and will have a certain conversion gain (depends on the type of mixer) and it will have an associated noise contribution.
On a standard radio this mixer gets presented with a wall of amplified thermal noise from the first rf amplifier.
This wall of noise will be above thermal noise by approx the amplifier gain plus its noise figure minus the losses in the bandpass filter after the amplifier..
So ballpark 17 to 20dB above thermal noise perhaps.
If you just look at the receiver at the input of the mixer stage it will have its own gain and noise figure. It also has a lot of lossy circuits after it including filters etc before the first IF amplifier stage (with assoc noise figure)
I think you will find on many older Uniden based CBs that the system noise figure looking into the mixer is quite high. Maybe 16dB?
So even if you could reduce the first RF amplifier noise by 2dB I doubt the sensitivity of the whole radio would go up 2dB because the inherent noise contribution at the mixer will become significant with respect to the lower noise of the front end.
I hope this makes sense. The way to design a receiver for a certain noise figure is to cascade the gains and losses of all stages of a receiver and enter them onto a spreadsheet that calculates the total system noise figure.
This is the figure that matters 
Because of this I believe it is going to be quite difficult to make a significant improvement in system noise figure in the centre of the band by just swapping out parts in the RF amplifier. The mixer and the losses after it start to become significant if you reduce the RF amplifier noise figure. So you don't get full benefit unless you increase the amplifier gain as well as reducing its NF.
On a standard radio the significance of this post mixer stage noise is masked to a degree by the wall of noise from the first (noisy?) RF amplifer.
Note that I don't recommend an external preamp although this would certainly be one way to reduce the system NF by several dB.
The reason is that a preamp will degrade the dynamic range of the radio (in terms of detecting small signals in the presence of large ones) so I dont think the benefits of lower NF justify a preamp. CBs already have poor dynamic range so preamps are not a good idea
