As a sibling comment notes, the first layer of spins is usually made of atoms, and arranged in a crystal, and when the temperature is high enough the crystal will melt and you will not have the nice layer of spin anymore. The second layer is too close to magic, but I guess they will need to be supported by atoms and they will melt too.
Also, I don't understand how this system is related to superconductivity as they say in the article. I see no Cooper pairs, and my understanding of high temperature superconductivity is too weak, but it's about something line flux pinning and it doesn't look related at all.
I've sent a lengthy email to Atomic Rockets strongly suggesting that cloaking devices are back on the menu. No reply yet and I know I've said this a few dozen times before but I really think we got 'em this time!
Heat would definitely kill fusion reactor magnets.
That’s why they’re so hard to build, plasma of a million degrees has to sit only a meters away from superconducting magnets which have to stay cryogenically cooled to near absolute zero.
Everything idealized is posdible on paper.
Simply put physics says if you put enough energy on an atom in a solid, it will eventually leave the solid on the solid/vacuum boundary from raw kinetic energy alone. Now, if you're trapped in an infinitely large solid, there would be no boundary and the magnetism could remain. For example the example of the early universe would be an example of one of these near infinite structures where magnetism remains. Large astronomical objects with lots of gravity may be able to reproduce these conditions somewhat like this.
With all that said, it should give us means of maintaining magnetism in an ordered manner in structures we'd consider hot as a human.