I searched the Irish Family History Foundation's database. It contains 11,500,000 surviving rcds for abt 60% of Irish counties. Rcds B4 1840 are sparse.
1. For Michael McGinnis (+variants) M 1838 +/- 4 yrs, I found 1 M rcd in 1838 in Lordship parish. The forename was written as "Michl." There are 2 Mary Hagan M rcds in Co. Louth in 1838. However, I couldn't find a matching M rcd for Mary Hagan in the same parish as Michl. Therefore, the 1838 M rcds MAY be false positives.
2. I couldn't find death rcds for this Michael McGinnis for any date in Louth. Of the three types of rcds, the death rcd was the least well kept.
3. For John McGinnis (+variants) B 1838 +/- 4 yrs, I found 2 B rcds for 1838: 1 in Dundalk parish, the other in Lordship. The one in Lordship leads me to speculate that the Michl M rcd in 1838 Lordship MAY be correct and the absence of a matching Hagan rcd is due to a transcription error. A view of the detailed rcd SHOULD show whether the parents are Michael and Mary.
4. I found 2 Mary Bellew M rcds in Louth: 1 - 1876, the other - 1879. Neither were in either Lordship or Drogheda parishes. There were no John McGinnis marriages in Louth after 1846. The marriage may have occurred elsewhere; Note-there was a great deal of social turmoil and internal displacement in this era due to Famine.
5. For a Mary Bellew B in 1842 +/- 4 yrs, I found 1 B rcd for 1843 in Drogheda. This is the only M rcd which fits your data. Only 4 of 20 parishes were NOT keeping rcds in this span. Therefore, this rcd is likely that of your ancestor.
6. Assuming that 1838 was the correct DOM, for John's siblings, I found evidence of at least two McGinnis families in Lordship parish. The childrens' names and DOBs are: Arthur 49, Bridget 39 & 40, Catherine 38 46 55, Elizabeth 43, Henry 39, Jas (James?) 45 & 46, John 46, Laur 40, Mary 41, and Thomas 38. All McGinnis' families apparently had left the parish by 1855: no more births after this date. The only way to segregate births by family is to view detailed rcds for each and sort them by parents' names.
7. The priest in Lordship parish used abbreviations for many forenames. Hence, Tho was used for Thomas, Mihl for Michael, Jas for James, Pat or Pat'k for Patrick, et al. When searching, it is best to use the 1st letter of the forename.
8. For Michael McGinnis (+variants) B 1813 +/- 4 yrs, I found only 1 B rcd. The forename is listed as "M." This may have been a priestly error, or the original rcd was in a state where the rest was illegible. Kilsaran is the parish, the date 1809. Only 6 of 20 parishes kept rcds at this date. Hence, it is unlikely that this rcd is for your ancestor.
9. Suggest you google the Foundation's name to lead you to the database. Replicate the searches above. Then, you can decide whether to continue a search of the detailed rcds.
I hope this gives you a few clues. Post a msg to Genforum to tell me your results.