Hello, I don't know if I'm on the right place to ask this question but I always wanted to know how did they did a thing like this: 
(4:41 to 5:18)
I don't know how they did put those image together and make a stepmotion like animation.
Thank you very much !!!
You probably are in the wrong place ... but there might be some here that do "animation" of this type .. To learn to do it, I'd suggest you download the animated video you reference and study it "frame by frame" in Vegas. For example there are roughly 20 still frames of the "eyeball" movement and "shocked look" that was done in the video at 5:03 to 5:04. How was this done? Likely in some photo editing software that could redraw over the image. Key parts of the face were redrawn to generate the effect over the course of the 20 or so frames. It's not that easy to do something like this right. There's a lot to learn about doing animation frame by frame. But, in a sense, all film or video is generated frame by frame and it is only our persistence of vision that makes it appear to be smooth when played back at the right speed.
If I understand the question correctly:
With Vegas it is quite simple: You have stills of the different cut-outs on different video tracks.
The stills may be in .png format with no background - or in .jpg or whatever format but on a green/blue background from where you can isolate the cut-outs by some chroma keying (green screening).
In the different tracks you can move the pics around, turn them and zoom in and out.
I think ritsmer got it right. This film was done circa 1971, just like old cartoons hand drawn and filmed. Thanks for linking to the film. I was a Zappa fan in the 60's and 70's. If I recall this was in his 200 Motels film.
Frame by frame on rostrum. 
is instructive.