In short, you asked a general question that is impossible to answer without knowing all of the details of your application. Only general observations can be made.
Most cooling tower applications utilize 6-8 cycles of concentration. That is an optimum range considering the cost of chemicals and blowdown requirements. The cost of cooling tower chemicals increases greatly when you decrease the cycles of concentration. The cost of the raw water and disposal of water have to be addressed.
If you start out with 500 mg/l TDS water quality, and use 7 cycles of concentration in the cooling tower, that would be 3,500 mg/l TDS in the cooling tower. If you only started out, with 250 mg/l TDS, that would be 1,750 mg/l in the cooling tower. Cooling tower manufacturers place specific operational limitations on water quality parameters such as TDS, SS, Calcium among others.
What you are dealing with is a systems type of problem. You have various inputs effecting the outcome. You may have different water quality parameters depending on the water source that may be more corrosive or more scaling. You can have your cooling tower built of more expensive materials that can handle the more corrosive water associated with higher concentrations of salt. You raised other factors such as water rights, available well water, and LEED credits that may offset the start up costs. The size of your system has a bearing on the recommended approach as well. There are lots of trade-offs.
Generally what is done in an application like yours is to reserve the better quality of water for the applications that need higher quality of water. Applications that can get by with lesser quality have to make do with the water quality that is available.
RO technology is easier to apply on a source of good water quality. I personally would not recommend RO to be applied on a cooling tower blowdown. If you are a power plant, my first choice would be an evaporator for desalination as shown in this project: