The answer is partly that this is an old technique from when little was known about the genome so enzymes were less good and using little enzyme for a long time was often needed for complete cutting which is essential for a Southern otherwise you get a smeared signal over a wide size range. Enzymes are now better so shorter cut times are good.

Similarly we had no idea what size fragment the probe would anneal to so electrophoresis conditions were set up to separate both large and small fragments as well as possible but if you know the fragment sizes then gel concentrations and run conditions should be changed to suite your experiments.

Stringency was controlled both by the salt concentration and temperature in the annealing stage but also by the wash conditions with high salt washes giving a strong signal but with higher background while low salt washes were good for washing away all background ( non specific signal) but also washed off some of the specific signal giving a weaker signal. When the probes were radioactively labelled it was common to do a high salt wash to get a strong dirty signal an prove it has worked then re wash in low salt to get a cleaner signal. This avoided the problem of a too stringent wash leaving no signal and you did not know where to look for the problem of having no signal