It is pretty nice, and the models get better every day. As a caveat, I must say that there are LOTS of things to consider when doing research with iPSCs, mainly related to how accurately your cells represent actual human cells. For example, I work with iPSC-derived neurons and any change in the process of maturation of iPSCs into neurons vastly changes the properties of the final cells. Also, neurons and other cell types take long to mature from iPSCs. In the case of my cells, it takes around two months until they are at a stage I can use for my experiments, and they have to be maintained for that long and lots of things can go wrong.

All of this is, of course, also a problem with ESCs, but not with animal models. If human genetics are important for your experiments, iPSC models are almost the only choice you have. If there is a good animal model for the disease you're studying and organism physiology is more important, then animal models are better.

If you mean embryonic stem cells, their main advantage over iPSCs is that ESCs have been around for longer and can be considered somewhat more reliable. iPSCs need to be generated in the first place from mature cells, and although this process is relatively simple (only four factors in the case of Yamanaka's Nobel prize-winning research), there is still a lot of debate over how it should be done and how it can affect the resulting phenotypes.

ESCs, on the other hand, are already naturally capable of generating tissues, so there is a larger likelihood that the resulting mature cells will resemble more the actual human ones. In studies in which the mutations of the diseases that are being studied are generated by genetic manipulation, ESCs are still preferred by many labs because of this reason.

But this is an issue for iPSCs just because it is still a very young technology, and huge advances are being made constantly that make ESCs less relevant. Being able to obtain cells directly from patients is a huge advantage as it allows to study diseases that have an important but not entirely known genetic component, like most neurological disorders.