A load average of 1.5 is not an issue if you have more than one CPU allocated to the VPS. If there is only one CPU allocated, then you are bottlenecking somewhere. The 'sar' tool is good for identifying the bottleneck.
It may take time to fill the caches. It seems you have a reasonably memory efficient load. Give the system some time, and memory usage should increase. If it does not increase, then you may be hitting kernel tuning limits. Allocating more memory than is required will have diminishing returns.
Unless you have a very large database for which all data is actively accessed allocating more memory to the database will not increase memory usage. Maximum memory usages should be not much more that the space used by the database's data files.
Allocating and using excessive memory for a virtual server may cause issues for other servers running on the same host. There is a limit after which all virtual servers will begin swapping leading to a significant decrease in performance. (I have seen Java Full GC times go from seconds to hours.)