The purpose of the study was to investigate Neural Networks in general and RecurrentrnNeural Network in particular. The investigation was to construct a model that shows therncomplex data processing of these Neural Networks and their functions in a simple way byrnusing Artificial Neural Networks. Therefore, in order to carry out this task, Perceptron,rnFeed Forward and Recurrent Neural Networks, and the basic concept of BiologicalrnNeural Networks, which are among the contents of this paper, are very important.rnFrom a biological point of view, Neural Networks are networks of biological neurons.rnBasically, they are networks of sensory, motor and associative neurons functionally, andrnaxon, cell body and dendrites structurally. Now, when an external or internal signalrnreaches the sensory neuron through the receptors, then it passes to the motor neuronrnthrough the associative neuron within the central nervous system. Since motor neuron hasrnconnection with the brain, then it immediately fires through the command that comesrnfrom the brain. This means that it gives response for that particular signal. In this paperrnwe extend these concepts to mathematical model by using the concepts of mathematics.rnFor instance, a Heaviside function is a simple mathematical model of Neural Network.rnOn the other hand, Recurrent Neural Network consists of the graphG=(N,A)and arnfamily of formal neuronsG=(V,E)-(X,Y,a,S) each associated to one of the Vertices IEVThisrnpaper also provides details about the types of Neural Networks present in human beingrnand also provides details about the Artificial Neural Networks. This paper also presentsrnsome Applications of Neural Networks