Applying it to dust and gas particles (massive enough for gravity to shape it into a spinning sphere) in a vacuum falling towards (on a stable enough orbit) a gas ball with 99.8% of the total mass in the system, itself in hydrostatic equilibrium with its own gravity pull through nuclear fusion of hydrogen into helium ... Well, we are not used to that kind of motion i.e. geometry on astronomical scales. But once you see what dictates the (near) stable motions of all those tiny particles the solution is simple: in the solar system the sun is nearest to all objects in the system, on average. Whatever is nearest to the Sun is the closest on average to all other elliptical moving objects. From Neptune's perspective Mercury is barely moving and oscillating so fast as to standing still. So, in the case of planets the underlying geometry is moving along (stable) elliptical (i.e. closed path) orbits (approx. Newtonian mechanics), applying our "straight" lines for "distance" is amiss here and can lead to confusion. I personally like to see it more - inspired through e.g. Kepler's imagination[0] - as a geometrical-dynamic system (Kepler orbits) more akin to music.