This is very typical for a NAV implementation on SQL Server.
NAV uses a fat client which does all the processing. It sends out very simple queries to SQL with the result that CPU utilisation is zip compared to ‘normal’ SQL systems.
Don’t worry to much about it and try to focus on Disk I/O and network bandwith.
If you have client machines that do large processing you might want to consider upgrading them.
Because of the way navision locks tables it is very easy when everybody is doing simular things for the transactions to become serialised, that is they happen one after the other rather than them all happening at the same time and this is especailly true for table writes. So what happens is the first transaction will lock a table (quite often the number series table) and the next process will then try and lock the table and when it can not it waits for the first process to finish before it starts.
That might be one of your problems right there. With the logs on the most active drive array, it is wonder it is slowing down. It is recommended that the transaction log is on its own dedicated array. This means nothing but the transaction log for the NAV production database. Put anything else on that drive, like backups, or transaction logs of other databases, and you adversely affect the performance of your NAV production database.