Georgina,
I would like restricting our consideration to what might be essential mistakes you have in common with Einstein. I like your thinking because it may guide me to neglected origins of Einstein's thoughts. Both you and Einstein's theory prefer the B-time which means ignoring the now. Perhaps it does not matter much that your reason for this preference is different from Einstein's.
Well, those like you who are considering only the present moment as existing but past as well as future configurations as non-existing cannot swallow that there is no observable state called present but a decisive ideal border in between them. Indeed, recoded histories as well as models are not bound to the in reality naturally given ideal reference point "now" between past and future. Your "causality front" between nothing and nothing sounds silly to me, because I see not only all past but also future processes being subject to causality.
Einstein just shared a mandatory up to now tenet: The laws of physics are invariant under shift of time. Formalizing of this option for shifting time shift goes back at least to René Descartes who, when he created his Cartesian coordinates, strived for non-arbitrariness while he de facto introduced the necessity to choose an arbitrarily agreed reference point t=0 with the option to shift it. Physicists seem obviously still having problems to admit the difference between reality at the basic level of abstraction and even the best matching model abstracted from it.
I quote from https://www.bartleby.com/173/7.html : "Of course we must refer the process of the propagation of light (and indeed every other process) to a rigid reference-body (co-ordinate system)".
Actually, Einstein's thought experiments of 1905 and 1917 dealt with two rigid bodies A and B that are thought to steadily move relatively to each other along a straight line. Doppler-effect causes the same blue-shift of signals sent from A to B and vice versa from B to A as long as the motion shortens the distance while it causes a red-shift for good after the reference points met each other. For symmetry reason, clocks at A and at B may be a priori assumed working in synchrony, and there is no twin paradox. The corresponding decreases and increases of length, respectively, are just apparent ones.
Must we indeed refer the process of the propagation of light to either an emitter E or a receiver R or a body-like medium M between A and B? No. On the contrary, Einstein correctly rejected ether theory and Newton's ballistic, so called emitter theory. They have no convincing logical justification. Lorentz' ether theory proved at odds with the experiments by Michelson and Michelson with Morley.
Nonetheless, there is no reason to question that the velocity v of light in empty space always equals to a constant value c if we calculate this c as simply the distance between E at the moment of emission and R at the moment of arrival divided by the time of flight between E and R. This excludes referring the light to either E, R, or M.
Calculating v as the mean value of propagation from A to B and return would be correct with v=0. However, application of such Poincaré/Einstein "synchronization" for v =|= 0 is the incorrect and unnecessary basis on which Einstein's whole theory of Relativity was fabricated. Einstein didn't provide any logical reason for this calculation in case of relative motion between A and B.
Let's now check what I quote from https://www.bartleby.com/173/5.html :
"If, relative to K, K' is a uniformly moving co-ordinate system..., then natural phenomena run their course with respect to K' according to exactly the same general laws as with respect to K".
Are these seemingly compelling principles already fit for use in general? No. General laws do not imply that variables are invariant under Doppler or other time shift. Well, it is common practice to arbitrarily choose any point as the reference. There is no a priori privileged one in space. However, once the reference has been chosen one must not refer to a different one. Moreover, elapsed time has already a non-arbitrary reference point, the now.
EB
