Exactly so. Emphasis on the "may" however, if you're looking for better temperatures its worth establishing if you need more cooling surface area or greater flow rate. If a water thermometer is unavailable then the following (inaccurate and subjective, but quick) tests may be worthwhile.
1/ If slowing the fans down/switching some off has little to no effect on temperatures, you wont benefit from another radiator.
2/ If decreasing flow rate, say putting a clamp on the tubing, has a dramatic effect on temperatures, this suggests you're flow rate limited.
3/ Temperature difference between cpu and gpu with comparison to a similar set up, if cpu load temps are disproportionately high this suggests flow (smaller die => impingement blocks, which are only effective at high pressures)
4/ Touch the radiators, and stick your hand in the airflow from them. If either is noticeably hot it suggests you'll benefit from another radiator. Ideally the water temp should be low enough that you can't really tell that the radiator is warmer than the room, and the airflow feels fairly cool.
Using my current loop as an example. I have an ek supreme, two gpu blocks, a chipset block, three radiators, and two quick disconnects. A few 90 degree barbs too. Adding a forth radiator had no effect whatsoever on temperatures, however going from one 18W DDC to two in series took ten degrees off my cpu temperature. GPU temperatures were unaffected by either another radiator, or by the second pump.
This loop is rather more restrictive than yours, but equally the pumps involved have higher head pressure so are better suited to multiple block loops. Flow rate limited systems rarely come up in computer watercooling, as multiple loops are fashionable and people tend to over-specify pumps. Loop order doesn't matter if flow rate is within normal bounds, but it will matter if you're flow rate limited as the temperature drop across a radiator will be higher.
If you can list components (processor, water blocks etc) it should be possible to deduce the temperatures you'd see at normal flow rate, and from that + your temps where the bottleneck is. Radiators (with known fans) have known C/W values, and waterblock testing tends to give a graph of cpu - water delta vs flow rate. Power consumption of processors/graphics cards/pumps is generally known. Its probably easier to stick a thermometer in the reservoir though.
