It's probably not a coincidence that yellow is both
(1) the color perceived as lightest, and
(2) one of the extremes along the yellow-blue axis.
The video mentioned that Long (L), Medium (M), and Short (S) are the preferred names of the types of cones sometimes called red, green, and blue, respectively.
The video doesn't go into the physiological basis of the opponent process, but there are neurons that form the sum and difference of signals from L and M cones, and other neurons that add/subtract the S signal to/from the sum of L and M.
Different neurons compute
the "yellow" sum Y = L + M,
the difference x = L - M,
the difference y = Y - S,
and the sum z = Y + S.
Subsequent processing in the brain involves only x, y, and z, where x specifies position along the red-green axis, y specifies position along the yellow-blue axis, and z is the luminance (i.e., apparent brightness).
It seems reasonable that for a given value of z, the spectral color most easily confused with white would have the highest value of y subject to the constraint that x = 0. This is a hypothesis which could be tested by a psychophysical experiment.
= new reply since forum marked as read