I don't have much experience of model railway designs, and had no real idea how much track would fit into what space. The choices I say were either to "try it and see" or try to get a stronger up-front design. I chose a CAD approach, because I'm used to similar tools for other things; it was a choice that was definitely right for me!
Why use a CAD design? With a good CAD tool, it is possible to understand how the design will work as a whole. You can visualise how all the track will look,and where it will be. You can find out how long the platforms in a station will be - or, conversely, find out where the points need to be for a given train length. You can find out whether that 5th track will fit into the fiddle yard and still allow the right train length to fit in.
For a multi-layered design there are other benefits. You can establish the height of the track at any point, and plan the gradients to be achievable for the rolling stock in use. You can visualise whether a point motor on an upper layer will foul the track on a lower layer. I was able to design the baseboards so that cross bracing timber was not going to obstruct point motors.
One of the key benefits of a CAD design is consistency: there is one design record rather than a lot of different ones, where differences could creep in. If I make a change to a track, then I can change the point numbers and power feed arrangements consistently in the same tool.
I was recommended a tool called XTrkCAD. This is a "freeware" tool that can be downloaded, and once registered (at zero cost) it offers unlimited operation. I have found little that XTrkCAD can't do that I wanted it to do, and I've been very pleased with the results. It supports libraries of track from most manufacturers including the complete Peco range. You can also drag simulated trains onto the track and operate them: this allows problems with train and platform lengths to be identified quickly.