I examined the logs a bit more closer and you're quite right it is quite a long time. I looked further into it and seems there is some backups getting backed up.
I think if you exclude folders similiar to the following you would cease to have issue with the backup running for a long time:
Those look to be backup locations and contain very large files. What is happening is the script is processing a large file, but isn't able to handle it in less than the 3 minutes maximum we set. If we obeyed your timeout of 15 minutes it would mean if a memory problem occurred and backup stopped, it wouldn't resume for 15 minutes, so we stick to the 3.
I'm not sure what the other systems are backing up but it must be skipping those folders maybe and only taking specific WordPress components, maybe even just the database, as 20 seconds seems a bit too quick been as our script in the log has taken more than that time on one single file in one of those folders.
See if excluding those locations will work for you. Meanwhile, we've been working on progressive backup for large files and better handling of detecting the 3 minute mid-way through file processing, which would work around all this issue you are having. But this is not due for a while.