There are neurons that signal fast, and neurons that signal slow.
There are neurons that send signals that quiet other neurons (they’re called inhibitory), and there are neurons that excite other neurons (they’re called excitatory). There are neurons that fire in pulses, and there are neurons that just turn their voltage up and down when they communicate. There are neurons that connect vastly different areas of the brain, and there are neurons that only connect in their local neighborhood. Scientists don’t even know how many types of neurons there are in the human brain!
Treating neurons as transistors is one of the simplest models researchers use. Rather than model neurons as on/off like transistors, more complicated models include having neurons that produce a firing rate, or a probability of firing, or even describing the membrane potential at particular locations. Researchers can adjust the level of abstraction in their model to try and describe something interesting that neurons do.
Different models capture different things. Some models capture that neurons communicate using voltage as an electrical signal, some capture that neurons generate that electrical signal with chemicals, and some don’t care how neurons do what they do so long as they fire.
Researchers also try to describe how neurons connect to one another. Some neurons connect only to nearby neurons, some reach far and wide. Others only connect to neurons that fire for a specific stimulus. There are so many different types of neurons in the brain, that researchers could spend forever defining rules for how one type of neuron connects to another type.
Statistician George Box is famous for saying “all models are wrong; some are useful."