Is Schrodinger's Cat itself an observer?

The point is, it has made you think about the issue. Whereas we all might agree a hydrogen atom is not an observer and a human is an observer, the case of a cat is not so clear. The point of the thought experiment is to expose problems with the Copenhagen interpretation - which it does very successfully.

This isn't actually a mystery and the answer is fairly simple: real quantum physics is equivalent to many worlds quantum physics. There really isn't an alternative explanation available.

Yes, I know there are a number of other interpretations available. But I was specific in my language. Copenhagen, for example, doesn't explain anything. It just says "this is what appears to happen and there is no explanation."

As it turns out, the other interpretations do this as well. Bohm, for example, 'explains' how there is actually no such thing as superpositions (by placing all the mathematics of a superposition in to the 'pilot wave' -- mathematically equivalent to superpositions but not called that) but doesn't even attempt to solve any of the big problems of quantum physics, such as your question (know as the observation problem) or how quantum physics can be squared with locality. Or put another way, Bohm 'explains' something that needed no explanation (superpositions--which are in fact real) but doesn't even attempt to explain / solve any of the problems of quantum physics.

This is true for every single 'interpretation' of quantum mechanics except one: many worlds.

Many worlds isn't merely another interpretation. It's an actual explanation of why quantum mechanics behaves so counter intuitively to us. And it does it by simply taking the theory seriously and refusing to compromise on it. It literally add nothing at all to the theory. It just says "well, if we take the theory seriously, what does that mean?" This is the complete opposite of every other 'interpretation' all of which add extra stuff to the theory that serves no purpose but to not take the original theory to it's own logical conclusions. That is why I say Many Worlds Quantum Physics is Quantum Physics. All the others are Quantum Physics plus some extra unnecessary stuff that just complicate the theory for no particular reason.

So how does many worlds quantum physics explain Schrodinger's Cat? When the quantum event takes place that both kills and doesn't kill the cat, reality superpositions and there is then a dead cat in one world and a live cat in another. When you walk over and open the box, you then superposition. One version of you in one world sees the dead cat and one version of you sees the live cat.

"Observation" plays no role in many worlds quantum physics except in the sense that it explains why observation seems to cause the wave function to collapse -- namely that you become part of the overall super position. In fact, it never collapses at all.

If you are curious to explore this answer further than I can explain here, look but the books by David Deutsch (creator of quantum computational theory). Both of his books, The Beginning of Infinity and The Fabric of Reality, explain this in much better detail.

Also, look up the Elitzur–Vaidman bomb tester experiment on Wikipedia and spend some time thinking about how many worlds would explain what is going on and see if any of the other interpretations even attempts to. It is much weirder than Schrodinger's Cat and much harder to explain away. For example, the wave function 'collapses' even when there is no observation at all! (Because the observation of the bomb exploding takes place in a different world. The world where you don't see the bomb explode never has any observation take place in 50% of the cases.)

Schrodinger used the hypothetical cat to illustrate what he thought was the absurdity of assuming that wave-function collapse only occurred when there was an 'observer' making a measurement. The point he was making was the common-sense notion that collapse of the wave function would occur when a particle interacted with any large body, whether it was a measuring device or not.

As for what constitutes 'an observer', an observation is nothing more than some interaction at a microscopic scale that has a quantitative consequence. When you observe an emission spectrum, your eye is registering light that has been split through a prism; in other words the effect of photons interacting with the electric field due to the distribution of atoms within the glass. The interaction between the photons and the atoms within the glass takes place regardless of whether anyone or anything is 'observing'- the prism of the spectrometer acts as the 'large body' you mention in your question.