How does an magnetic domain align?

[Arceus74]

I have these assumptions please correct me if they are wrong:)
1)A particle has mass,electric charge and magnetic moment.
2)Somehow in a atom the net magnetic moment of all the particles including protons and electrons get canceled out and becomes neutral.However for half-filled electron shells there is some magnetic moment which makes the atom magnetically positive.
3)That is why some of the D block elements are magnetic.
4)When external magnetic field is applied the domains align.
In that case if we apply a magnetic field and the domains align it means that microscopically the electrons orientation or direction also changes,which makes the magnetic moment unbalanced right?
Please explain how does the magnetic moment is changing?

 

[Charles Link]

There is a very strong exchange interaction which causes neighboring spins to often be already aligned. This can create a cluster of atoms aligned with aligned spins which is a domain. The magnetization of individual domains can point in somewhat random directions with no externally applied field, and the macroscopic effect is that the material is unmagnetized. Upon applying an external magnetic field, many of the domains can align themselves with the applied field, (which is energetically favorable since the energy $E=-\mu \cdot B$), and the material is strongly magnetized. In some materials, when a high level of magnetization occurs, the external field can be removed and essentially the internal magnetic field (which can be computed from the bound magnetic surface currents) maintains the alignment off the spins and a permanent magnet results. For reasons that are somewhat obscure, in soft iron, a permanent magnet does not occur, but instead, when the external field is removed, the energetically preferred state is one of a bunch of somewhat randomly oriented domains, with the net macroscopic magnetization being small or near zero. Perhaps others can add to and/or correct my explanation. The magnetism is a rather complex subject, and a couple of the more authoritative books on the subject can get so mathematically detailed that the explanations they offer are difficult to follow in detail.