It's not a misleading analogy. It's not even an "analogy". It's the neurological fact. Not only is every neuron like a threshold gate, it is a threshold gate.
These gates work in very different ways, but they are nonetheless computationally equivalent. And of course, it is possible to think of an indefinite variety of other ways of making a primitive gate. For example Ned Block use's the example of Cat and Mouse to make an AND gate.
We are creatures that have a useful and interesting biological level of description which the computer model of mind ignores and aims for a level of description which aims for a description unconcerned with the biological realizations of cognitive structures. As far as the computer model goes, it does not matter whether our gates are realized in gray matter, switches, or cats and mice. Of course, this is not to say that the computer model is in any way incompatible with a biological approach
Whether or not we decide that something is a computer depends on whether we choose to interpret states in a certain way and whether we do so is up to us, we could assign computational function to many things that are big or complex enough to have enough states, such as a hypothetical system of millions of cats and mice. To quote Searle "We can't, on the one hand, say that anything is a digital computer if we can assign a syntax to it, and then suppose there is a factual question intrinsic to its physical operation whether or not a natural system such as the brain is a digital