Here comes the most important shot of all. The one thing that "sells" the reality of the situation.
We've got Aeetes (and goons) in the background, the animation in the middle ground, and Jason & Co. in the foreground.
This may be an actual process shot in which the foreground elements are composited. They may have even been composited in camera -note the hard black shadows -negative space making a bipack easier.
The foreground also has a different vanishing point from the background. That's something common in animation which doesn't have when you shoot through a lens.
Harryhausen may have done this shot in some other ingenious way. The important thing to remember -if you're going to do something more complicated, more expensive, make sure it's critical to the success of the film.
The "three layered" shot comes a point where we need to have the geography reinforced at the moment we need to be sold the animation exists in the same reality as the live action.
This following cut, is back to the skeletons approaching from the side with Aeetes in rear projection.
There should be no question to the geography now. We know what Jason sees, we know the relationship of his position to the skeletons.
We also know he's outnumbered 3 to 1.
Next up: The fight begins (almost)