Neurons

There are lots of different models of biological neurons.
We use Leaky Integrate and Fire models, and Fatiguing Leaky Integrate and Fire models.
Neurons integrate activity from other neurons, and if they get enough activity, they fire and send activity to other neurons.
There are lots of other models, and more sophisticated models (like Hodgkin-Huxley).
These are analog models, but we work mostly with 10ms and 1ms discrete models. They update every 10ms.
If you plug these together the right way, you can program anything.
Neurons connect to other neurons by synapses, that can be excitatory or inhibitory.
What's really interesting is learning with them. That is, the synapses are plastic.
There also might be something really interesting about distributed processing with them.