Basically, I just want to know that why we design different tanks for storing the main engine fuel and the RCS thrusters fuel. If we are using the same fuel for both the main engine and RCS thrusters (for example N2O4/MMH) then wouldn't it be structurally feasible to design a single tank for holding both fuels.
One reason for separating the RCS and main tankage is the ullage problem; to maintain good flow into the engine inlets, you need to separate the remaining propellant from the pressurant gas in the tank and ensure that the propellant is at the correct end of the tank.
As described in an answer to this related question, this is commonly done with either a flexible diaphragm between the pressurant and the propellant, or a wicking system. It's easier and more efficient to use such techniques on a small portion of tankage to enable the RCS thrusters to fire (i.e. perform an "ullage burn") and then use the acceleration from the RCS thrusters to settle the propellant in the main tankage.
For craft that have higher thrust needs or where cryogenic propellants become possible the performance difference between long storage life mono or bi propellants and shorter life types start to justify more complex systems (example space shuttle mix of solids, Hydrogen/Oxygen and hydrazine/Nitrogen Tetroxide). For some sample numbers see or the book Ignition by John Clark which covers the lack of a single perfect propellant system at length.
Propellants will suffer from several of the problems listed, and the craft design will have to accept a combination that are least worst:
Freezes during flight (basic Hydrazine)
evaporates during flight (Oxygen/Hydrogen)
Low density requiring physically larger tanks (Hydrogen)
Dissolves storage tanks and fixtures (Fluorine)
Toxic during handling or accidents (most of them)
Expensive (Boron/some exotic hydrocarbons)
Breaks down in storage (Nitric acid,peroxide)
Sensitive to contamination/handling (most mono propellants)
Hard to re-start (non hyperogolic combinations)
Where the OP asks "why can't we..." it indicates a starting assumption that "this doesn't happen so far". Whilst the other answers to this question have suggested valid reasons why there could be separate propulsion systems for RCS vs main translational thrust it is in no sense a general rule, or even typical.
For satellites up to at least 7 tonnes with all kinds of needs for translations in different axes and slow and quick reaction rates it is quite typical to use a single propulsion system for all satelilte functions.
The main class of satellites that exhibits this design choice are geostationary communications satellite and these undergo the main deployment manoeuvres, RCS to support the main deployment, North-South and East-West station keeping (i.e. lower thrust translation) and various attitude related functions, off-loading and emergency sun reacquisition all using smaller thrusters.
The usual choice of propellant/gas separation is a surface tension based propellant management device within the tanks which is capable of supplying RCS thrusters (say 10N) and the main deployment engine (say 500N) together.
A reason why this approach might not be used is where the mass of the satellite / sizes of the main thruster vs. small thrusters are such that the surface tension solution is more difficult, though I suspect it would be hard to justify a much larger engine (say 2kN) for an interplanetary probe of a few tonnes in the first place.
Plenty of recent interplanetary probes (Juno off the top of my head) have used this (500N main engine) too.