The point here is that wires and other anomalies cannot be evaluated in quick tests. A quick test aims to detect small changes in the FR and distortion factor as accurately as possible, and this is where its advantages end. Long tests (a few minutes and more) evaluate everything else, revealing those little subjective things that turn sounds into music.
Listen as long as you want. It make no difference. Except the longer you listen, the harder it will be to identify a difference. If you wish to hamper yourself that way, that is your choice. I would suggest doing both.
A quick test aims to detect small changes in the FR and distortion factor as accurately as possible
No, this is not correct. Switching times are typically short because our echoic memory is very short (<5 seconds), and our working memory is also very short, i.e. <30 seconds, and comparisons using these memories are far higher resolution than any other memory imprint. Once you go much beyond 30 seconds, effectively you are not really comparing two sound anymore, but are comparing factors you have identified (or not) in those sounds, hence why the reliability of detection of changes drops.