Often, this involves a fan running at reduced speed

Try wiring two "whisper" fans in series. Even quieter than a whisper :-)

Common sense tells me that anything which moves air over the chassis will be
much more effective than some marginal change in the covering of tube
shields.

Phil Nelson

Thank you guys!

Colin:
Heat still remains in the cabinet.

True.I find rather amazing how many old design types, esp. for consumer
tube gear, were oblivious to heat. I think ppl had a "machinery" or
"engine" paradigm in mind. If stove and charcoal iron are hot, why not
a radio?

WRT boatanchors, I always tried to improve ventilation, with a
suspicion that in so doing I might move frequency-defining parts off
the expected temperature range, but the '390 seems to me quite well
temperature compensated. When Collins folks were designing this, bitter
cold in Korea must have been a fresh memory.

Right now that radio is coverless, perched mid-air on a rack-on-wheels.
I run a room fan aimed at it, and the difference in temperature is very
remarkable. The previous owner, I2FZX (sk) used it for decades with
covers on, sitting almost directly on a surface, with an HRO-500 right
on top, and never had problems with it. Maybe the HRO did.

Longer term, I plan to put it in a desktop wooden cabinet, stacked with
some SSB/synchro-AM device and a speaker (and a thermometer??), with
plenty of forced airflow. I am considering flowing air from the bottom
- out the lower side holes - into the upper side holes - out the top -
sucked out the back by a bank of slowed-down, underfed 120mm fans. I
also want to do the ss-rectifier mod, and introduce a NTC resistor for
inrush current limiting, both of which will cut heat.
_____

Back to the original question, I'll do some experiments with engine
paint and a meat thermometer. If it's good for telling if meat is
cooked, it may help avoid cooking tubes. :-)