Step 4:  

clear;
clc;
close all;

mynetwork=feedforwardnet([7,7]);  % Create the network
mynetwork.layers{1:2}.transferFcn= 'logsig';  % Change the activation functions for the hidden layers to logsig
mynetwork.layers{3}.transferFcn= 'tansig'; % Change the activation function for the output layer to tansig
mynetwork.trainFcn = 'traingd';

Set the learning rate:

mynetwork.trainparam.lr = 0.2;  

You have now completed creating the network and changing a few of its parameters.  Run the program, then type view(mynetwork) in the command window to also see its structure.   Also note that this network is ready to take data.  The number of input and output nodes for the network depend on your data, which is why they are shown as zero in the network diagram (when you use "view(mynetwork)").