What do you think?
I think it goes to show that some manufacturers don't give good advice about their
products. Not surprising, given the amount of "lore" floating around that was once
considered state-of-the-art, but really is rather mistaken.
You want the choke as close as possible to the potential source of current on the
feedline because that makes it most effective - that usually means near the
feedpoint. In the case of the 75 ohm matching section on the HF6V, putting the
balun between the matching section and the antenna may change the match,
depending on the length of coax inside the balun. In that case I'd put it at the
end of the matching section (though you could wind the matching section around
a suitable ferrite core and make a balun that way.)
Here is an actual example: at the County Emergency Operations Center we have
an Off-Center Fed Dipole for 40m and 80m, and it was picking up a lot of noise
from the computer network - about S9, making it unusable much of the time. We
decided to try adding choke baluns to eliminate the common mode current on the
coax, which was causing the coax to act as an antenna as it ran along the network
cables through the building. We started by putting a choke near the transmitter
in the shack, and it made no difference. Then we got up on the roof and added
one at a convenient splice and the noise dropped to S4 - S5, a significant
improvement. We still left the one in the shack as well, just for good measure, but
the one on the roof (as close as we could get to the antenna) certainly made
the most difference.