Cosmology is roughly speaking, the scientific study of the history of the universe. This is a worthy pursuit, but only to a point.
Right now, the universe has certain laws that it obeys. Nothing moves faster than the speed of light. The universe is expanding in a manner consistent with a simple cosmological constant. General relativity governs the gravity and describes the nature of space-time. Baryon number and lepton number are almost perfectly preserved as separate quantities. Mass-energy is conserved. The quantum physical laws of the universe obey CPT symmetry, even though they are neither CP symmetric nor T symmetric. Entropy increases over time. Baryon number conservation and lepton number conservation severely limit the creation of antimatter. The universe is predominantly made of matter and not antimatter. There are invariant physical laws whose physical constants and physical laws in the Standard Model and General Relativity do not change.
Extrapolating these rules of physics back in time can take you a very long way. It can carry you through the formation of all of the atoms in the universe. It can take you back to before the "radiation era" more than thirteen billion years ago. It can take you back to a point in time where the mass-energy in the universe was extremely smoothly distributed in a universe that fills a far smaller volume than it does today and the ambient temperature in the universe was close to the GUT (grand unified theory) scale where all of the forces of nature start to look very similar to each other.
There are questions, however, that one cannot answer by simply extrapolating back the rules of physics without making up new ones. You can't answer the question, "why do we have precisely the amount of mass-energy in the universe that we do?" You can't answer the question, "why is the universe mostly matter and not antimatter?" You can't come up with a principled answer to the question of how our current baryon number and lepton number in the universe came to be what it is today. You can't answer the question of why the physical constants are what they are today. You have to violate laws of physics like the speed of light limitation to get the universe to be sufficiently smooth in the first second or two of the universe.
Rolling back the clock, at most can give you a set of initial conditions. At proper time T, when the universe was X meters across, the laws of physics and physical constants were what they are today, there was this much mass-energy in the universe, the baryon number and lepton number of the universe respectively were Y and Z, the universe was A% antimatter and O% ordinary matter, and so on and so forth.
It is conceivable that this extrapolation backwards in time may even make it possible to get back to the first few minutes, or even seconds of the universe. But, from decades of trying we have learned that there are questions that can't be answered simply by extrapolating back in time with the existing laws of physics. But, there are limits that can't be explained without new physics.
My own bias and prejudice is to stop when we reach those limits. Cosmology should legitimately take us back as far as possible using the existing laws of nature to a set of initial conditions that had to exist that far back in time. This is a very sensible place to call "the beginning" from the point of view of scientific cosmology. Indeed, the initial conditions themselves may be suggestive of possible new physics that could bring them about. But, at that point, we start to engage in the process of scientific mythmaking, and stop engaging in the process of science itself.
Given that we have a thirteen billion plus year Big Bang cosmology that can't take us back before a singularity at t=0 in any case, who cares if we choose to start counting at t=0 or t=two seconds or t=ten minutes or t=one week or t=100,000 years. As long as we go back as far as we can with existing laws of physics and set initial conditions for that point in time, any early initial conditions that require new physics is just question begging. If you have to begin somewhat, why not choose a point of beginnning that goes back as far as your expertise can support, but no further.
This may mean that we never get a satisfactory answer to some of these questions, but so what. We will know what is important and will have a conclusion around which a scientific consensus can be built. If that means leaveing the source of those initial conditions unknown and unnatural, then so be it. Life doesn't promise us answers to every question.