Low-grade magnesium oxide (LG-MgO) was proposed as ordinary Portland cement (OPC) or lime substitute (CaO) for metal(loid)s remediation in contaminated soils. Some metal(loid)s precipitate at pH ≈ 9 in insoluble hydroxide form thus avoiding their leaching. LG-MgO avoids the re-dissolution of certain metal(loid)s at 9.0 < pH < 11.0 (pH-dependents), whose solubility depends on the pH. A highly contaminated soil with heavy metal(loid)s was stabilized using different LG-MgO by-products sources as stabilizing agents. Two of the three studied LG-MgOs were selected for the stabilization, by mixing 5, 10, and 15 wt.%. The effect of using LG-MgO not only depends on the size of the particles, but also on those impurities that are present in the LG-MgO samples. Particle size distribution, X-ray fluorescence (XRF), X-ray diffraction (XRD), thermogravimetric analysis, citric acid test, specific surface, bulk density, acid neutralization capacity, batch leaching tests (BLTs), and percolation column tests (PCTs) were techniques used to deeply characterize the different LG-MgO and the contaminated and remediated soils. The remediation’s results efficacy indicated that when the medium pH was between 9.0 and 11.0, the concentration of pH-dependent metal(loid)s decreases significantly. Although around 15 wt.% of a stabilizing agent was appropriate for the soil remediation to ensure an alkali reservoir that maintains optimal stabilization conditions for a long period, 5 wt.% of LG-MgO was enough to remedy the contaminated soil. When evaluating a polluted and decontaminated soil, both BLTs and PCTs should be complementary procedures.