You can't have two overlaying axes with the same x value and use bar with only a single variable because then the bars will always be at the same location on the x axis. I suppose one could get very clever about setting the xlim values on the two, but that would be tricky at best...
The solution to your dilemma is the magic of NaN -- create a dummy variable for each of the two variables in turn and plot the both as if they were grouped as your last plot --
x=datetime(2022,11,24:28).';
y1=randi(20,size(x));
y2=randi(500,size(x));
hB1=bar(x,[ y1 nan(size(x))]);
yyaxis right
hB2=bar(x,[nan(size(x)) y2]);
But, while this does what you wanted, I'd contend it is misleading owing to the scaling on separate axes makes the two variables appear to be nearly the same in amplitude.
Sometimes, "just because you CAN do something doesn't mean you SHOULD!"
