[Arran]

OK, yesterday I res-stuffed the first tubular guy and took out the twist-lock cap.
The twist-lock, as it's been the case with the previous Zenith cap, in fact shows that it is not dry. It should be 5uF, and two 15uF caps, 300V and 250 and 450V. The capacitances are 8uF, 18uF and 30uF. Considering it is 40-s, and not 30-s.....
I am tempted to reform it and see how it behaves.

I need to order a bunch of the caps, like 0.1uF. I could use 0.15uF I have but as the sch is barely readable when it comes to ref designators I cannot quite see which cap goes where and so though I am sure in some places they can be used, in some others it is possible 0.1uF is preferred. If only I could have a nice readable schematic with ref designators showing clearly. Oh well, can't have it all at once.
On  the other hand, this murkiness is more pronounced on the Beitman's and only in printed form, the first PDF Nathan gave is kinda readable. And yes, some of those caps should be what they are although 0.15uF would work too. Like those in teh detector circuit in the feedback bypass.

Mike;
  I've run across that myself with the old twist lock style condenser cans, I once repaired a DeForest-Crosley of the same age as your Zenith, as well as a few Canadian RCA and G.E sets with the same, and whilst they appeared to be good since the set still played with them in place I don't trust them. If they were from the 1960s or 70s I might roll the dice on reforming, but not on ones made before Pearl Harbor.
 I would just order some .1 uf 630vs, it's a standard size so any amount you order you will end up using at a later date on something else. That's one of the sizes I can confidently order in quantity to get a bulk discount, much like .01s, .005s, and .047s, it's like 90% of the tube type radios, TVs, and amps use at least some of them.
Regards
Arran