Quantum mechanics of the electron in an atom vs. Gryzinski's free-fall atomic model According to Heisenberg's Principle of uncertainty, you can not know the place and speed of a particle, at the same time. You only have probabilities of the estimate values.
These probabilities forms a region where an electron can be found, according to previous experiments.
These regions are these one represented in this image:

So, we know the duality wave-particle of the electron, but the Gryzinski's free-fall atomic model describes these shapes PERFECTLY. Is the Gryzinski's model right? It was introduced in 1965, so the duality of the particles were already known at the time. Can someone enlighten me on this subject?  
Video explaining the atomic free-fall model:
http://www.youtube.com/watch?v=P2IsIkSn5bk
 A: So it seems that Gryzinski's model is the same as the Bohr model, but adds a magnetic interaction and the result is that as electrons 'free fall' towards the nucleus there is a magnetic interaction which causes the electrons to return to their original separation from the nucleus. 
It seems to me that the Gryzinski model is, thus, treating electrons as classical particles and so it is not 'right'. 
... but then is any model perfect? 
The youtube video you show is very impressive - fantastic graphics, and it may be that something can be learnt from these calculations - the whole idea of it is pretty amazing for me and a wonderful extension of the bohr model. The only issue, though, is because it does not treat the electrons quantum mechanically it will have issues.....
...there is some research where useful calculations can be made treating electrons classically, but these are extreme situations normally, e.g. high power laser photoionization if I remember correctly.
Finally, I want to address your quest to find out if the theory is 'right'... Bohr theory is incorrect, but it does make some useful predictions such as the size of the hydrogen atom. I expect that Gryzinski's model can give some more insight, but it will ultimately fail as the electrons are treated classically. So most theories have some successes and some failures. It is helpful to know what level of theory you need - e.g. for NASA rocket trajectory calculations classical mechanics is probably fine, although it is not 'right' as it does not treat rockets quantum mechanically.
A: Those pictures do not mean what you think they mean. Specifically, if something is in the ground state, there is a nonzero rate at which you will find it much farther away than that sphere if you repeatedly do a position interaction on an ensemble all in the ground state.
So that sphere isn't a region in which a ground state electron lives just waiting to be observed.
Which is a problem with your characterization of the uncertainty principle. You write

you can not know the place and speed of a particle, at the same time. You only have probabilities of the estimate values.

If you can't know something, why would it make sense to that it even has a value? Even postulating that objects have positions and momentum's and that position interactions merely reveal the preexisting position and that mkmentum interactions merely reveal previously existing momentum leads to problems. Leads to predictions that disagree with observations.
Fundamentally, the order in which you do interactions changes the results you get, so you aren't merely passively revealing things, you are changing things.

These probabilities forms a region where an electron can be found, according to previous experiments.

No. A nonrelativistic wavefunction isn't a function of physical space, it is a function of configuration space. And the pictures you draw make it look like an electron won't be found far away. They can. Your pictures are just trying to describe regions that are somewhat more likely than others to give particular results. And in a very misleading way.
So if this other theory predicts that the electrons won't be found far away, then the theory is wrong.
A: Since Gryzinski free-fall atomic model is not mainstream I feel free to add a thought of mine. I'm coming from Complex one-dimensional structure of space, where a had postulated


*

*that dipole fields (magnetic as well as electric fields) have to have a inner structure and consist of two quanta and clusters from them

*this quanta are part of electrons and protons and added to charged particles during absorption of photons and re-emissioned during emission of photons.
The conclusion was that this fields require space and due to the continuous behavior of the field lines the approach of an electron to the nucleus stopps.


For a long time I thought that was it, but later take attention to the magnetic dipole moments of th electron and came to the conclusion that s- and p-orbitals arose from the mathematical apparatus of a harmonic oscillator in three dimensions and this useless tool for multi-electron systems prevents a deeper understanding of electron distribution around nucleas.
There is a better way to arrange 8 electrons (after the He-shell) around the nucleus, what I explained in About the distribution of electrons dipole moments in atoms. Only for two and for 8 magnetic dipols it is possible to arrange them in a perfect symmetry around a centre.

As I saw than this matches perfect the sp-, sp2- and sp3-hybridisation in chemistry.
