I owe you a few more anwsers, IMF. First, the hot rocks. I have already talked how some PIs may be set to shade out hot rocks and soft steels such as nails and simmilar. Obviously, this is not the way you would want to look for iron relics. Neither it is the way to search deep large targets, unfortunately. You wold probably want maximum depth from your machine, right? Under such circumstances a large lump of magnetite (or hot rock) is a legite target, it will be reported by the detector, there is no way around that. Yes, you may dig for nothing, you may also dig for nothing if your find turns out to be an old engine radiator or a modern anchor....
Second, would i invest more money or dig? I would definetly not invest more money over a single case - this turns out to be conterproductive - always. Build your hardware carefoully and grow with it. Would i dig - it depends. I will explain more of this later.
Third, if you are looking for a buerried or sunken ship made of steel, the magnetometer or more exactly the magnetic gradiometer is your best friend. You can buy something fency and expencieve or opt for a chip DIY kit where the only thing you will need to do is to connect sensors with provided cable and properly align them. You may find such kits if you google for fluxgate sensors, along with extensive tutorials. You may find usefoul to install the two fluxgate sensors on your boat.
When looking for a big deep target with a PI, first thing you should asses is to find out the signal comes from a deep large target rather than a bunch of small shallow ones. This can be done by using different settings and coils. Somethimes, some small targets should be dug out to properly verify the large and deep one. This is the first information you will get from your PI. Second, you should asses the depth and size of the target. Again, different settings, different coils and in depth knowledge of your machine will be necessary. This is a second information you will get from your PI. The third information you may get from your PI is optional; if you are using something really expencieve or if you have an extra magnetometer with you, you may find out your target is ferrous or non ferrous. This is all you may get from afortable and DIY machines. This is where casual metal detecting and deep seeking differs - you will be forcet to gether the missing info from other (historical) sources. You will not survive long without the proper investigation for every site you decide to research.
You may invert the time line of events, as i offently do, to make it look more casual, by first detect the desired sits and properly map the deep finds (i use GPS, camera, measuring band, ruller and precision #geological" compass to keep the booklet of "finds") and than investigate each one from other sources such as old writeings, maps, witnesess etc. At the end of the day you will be able to dig only a few of them and the better the information gathered, the less chance to dig for nothing.