Peratt claims that his computer simulations produce images that look like spiral galaxies. The problem is that his images are of the plasma distribution and he compares then to optical images. Spiral galaxies look like spirals primarily because the arms are bright since they contain star forming regions. Peratt's images have nothing in between the arms. In reality, the density of matter in between the arms is 10-20% less than the density of matter in the arms (not 100%).
Peratt claims that the his simulation of the plasma distribution of double-lobed radio galaxies matches their radio emission. But once again he is comparing different things. Double-lobed radio galaxies are actually hosted in elliptical galaxies.
At this point his model is invalid since it does not match reality. But there is more:
Peratt implicitly claims that gravity can be neglected in his simulation (since the code does not include gravity!).
But in the paper he authors with Meierovich (Equilibrium of Intergalactic Currents (PDF), B. E. Meierovich and A. L. Peratt, IEEE Trans. Plasma Sci. 20, p.891, 1992), equation 12 shows that the gravitational contribution to the energy balance equation is 7 OOM greater than that of the plasma contribution (just plug their numbers in!)
Peratt claims that galactic plasma filaments estimated to have a width of 35 kiloparsecs (100,000 light years) and a length of from 35 megaparasec to 3.5 gigaparasec (an average length of 350 megaparasec or 1 billion light years) exist. But the large electric current through them will cause synchrotron radiation. There is no evidence for this. There is also the question about the movement of filaments through the intergalactic medium which might cause shock waves and detectable X-rays. Add in the problem of why the filaments are not seen in studies of the mass distribution of matter within galactic clusters using gravitational lensing.
A last couple of points. These galactic plasma filaments will cross each other. So what happens to a galaxy if its filaments short circuit? Galaxies pass each other and cannibalize each other. How do filaments magically stick to their assigned galaxies?